{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "from fastai import *\n", "from fastai.vision import *\n", "from fastai.utils.ipython import *\n", "from fastai.callbacks.tracker import SaveModelCallback\n", "from sklearn.model_selection import StratifiedKFold\n", "import matplotlib.pyplot as plt\n", "from functools import partial" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "%reload_ext autoreload\n", "%autoreload 2\n", "%matplotlib inline" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[PosixPath('../../../Dataset/Herlev Dataset/best-rn34-herlev-multiclass-5fold.pkl'),\n", " PosixPath('../../../Dataset/Herlev Dataset/best-vgg19-herlev-multiclass-5fold.pkl'),\n", " PosixPath('../../../Dataset/Herlev Dataset/best-effb3-herlev-multiclass.pkl'),\n", " PosixPath('../../../Dataset/Herlev Dataset/abnormal_moderate-dysplastic'),\n", " PosixPath('../../../Dataset/Herlev Dataset/normal_superficiel'),\n", " PosixPath('../../../Dataset/Herlev Dataset/abnormal_light-dysplastic'),\n", " PosixPath('../../../Dataset/Herlev Dataset/abnormal_severe-dysplastic'),\n", " PosixPath('../../../Dataset/Herlev Dataset/normal_columnar'),\n", " PosixPath('../../../Dataset/Herlev Dataset/normal_intermediate'),\n", " PosixPath('../../../Dataset/Herlev Dataset/abnormal_carcinoma-in-situ'),\n", " PosixPath('../../../Dataset/Herlev Dataset/models')]" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "path = Path(\".\")\n", "data_path = path / \"..\" / \"..\" / \"..\" / \"Dataset\" / \"Herlev Dataset\"\n", "data_path.ls()" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "LabelLists;\n", "\n", "Train: LabelList (917 items)\n", "x: ImageList\n", "Image (3, 83, 146),Image (3, 106, 116),Image (3, 129, 119),Image (3, 108, 110),Image (3, 209, 173)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Valid: LabelList (0 items)\n", "x: ImageList\n", "\n", "y: CategoryList\n", "\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Test: None" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data_init = (ImageList.from_folder(data_path)\n", " .split_none()\n", " .label_from_folder())\n", "data_init" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "StratifiedKFold(n_splits=5, random_state=0, shuffle=True)" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "skf = StratifiedKFold(n_splits=5, shuffle=True, random_state=0)\n", "skf" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "tfms = get_transforms(flip_vert=True, max_warp=0.0, max_rotate=30.0)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "idxs = [[train_idx, val_idx] for train_idx, val_idx in skf.split(data_init.x.items, data_init.y.items)]" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "def model_callback(model, model_name):\n", " return [SaveModelCallback(model, every=\"improvement\", monitor=\"accuracy\", name=model_name)]" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[ .new_type at 0x7f601003a378>>,\n", " Precision(average='macro', pos_label=1, eps=1e-09),\n", " Recall(average='macro', pos_label=1, eps=1e-09),\n", " FBeta(average='macro', pos_label=1, eps=1e-09, beta=2),\n", " KappaScore(weights='quadratic')]" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "our_metrics = [accuracy, Precision(average=\"macro\"), Recall(average=\"macro\"), FBeta(average=\"macro\"), KappaScore(weights=\"quadratic\")]\n", "our_metrics" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Fold-1" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "ImageDataBunch;\n", "\n", "Train: LabelList (733 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Valid: LabelList (184 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Test: None" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "fold_idxs = idxs[0]\n", "fold_data = (ImageList.from_folder(data_path)\n", " .split_by_idxs(fold_idxs[0], fold_idxs[1])\n", " .label_from_folder()\n", " .transform(tfms, size=64)\n", " .databunch(bs=64)\n", " .normalize(imagenet_stats))\n", "fold_data" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Downloading: \"https://download.pytorch.org/models/resnet101-5d3b4d8f.pth\" to /home/fadege/.cache/torch/checkpoints/resnet101-5d3b4d8f.pth\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "1d51bd133eae43ae940b5cf4a502e254", "version_major": 2, "version_minor": 0 }, "text/plain": [ "HBox(children=(FloatProgress(value=0.0, max=178728960.0), HTML(value='')))" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "\n" ] }, { "data": { "text/plain": [ "Learner(data=ImageDataBunch;\n", "\n", "Train: LabelList (733 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Valid: LabelList (184 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Test: None, model=Sequential(\n", " (0): Sequential(\n", " (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n", " (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (2): ReLU(inplace=True)\n", " (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n", " (4): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " (downsample): Sequential(\n", " (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " )\n", " (5): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " (downsample): Sequential(\n", " (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (3): Bottleneck(\n", " (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " )\n", " (6): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " (downsample): Sequential(\n", " (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (3): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (4): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (5): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (6): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (7): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (8): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (9): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (10): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (11): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (12): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (13): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (14): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (15): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (16): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (17): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (18): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (19): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (20): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (21): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (22): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " )\n", " (7): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " (downsample): Sequential(\n", " (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " )\n", " )\n", " (1): Sequential(\n", " (0): AdaptiveConcatPool2d(\n", " (ap): AdaptiveAvgPool2d(output_size=1)\n", " (mp): AdaptiveMaxPool2d(output_size=1)\n", " )\n", " (1): Flatten()\n", " (2): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (3): Dropout(p=0.25, inplace=False)\n", " (4): Linear(in_features=4096, out_features=512, bias=True)\n", " (5): ReLU(inplace=True)\n", " (6): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (7): Dropout(p=0.5, inplace=False)\n", " (8): Linear(in_features=512, out_features=7, bias=True)\n", " )\n", "), opt_func=functools.partial(, betas=(0.9, 0.99)), loss_func=FlattenedLoss of CrossEntropyLoss(), metrics=[, Precision(average='macro', pos_label=1, eps=1e-09), Recall(average='macro', pos_label=1, eps=1e-09), FBeta(average='macro', pos_label=1, eps=1e-09, beta=2), KappaScore(weights='quadratic')], true_wd=True, bn_wd=True, wd=0.01, train_bn=True, path=PosixPath('../../../Dataset/Herlev Dataset'), model_dir='models', callback_fns=[functools.partial(, add_time=True, silent=False)], callbacks=[MixedPrecision\n", "learn: Learner(data=ImageDataBunch;\n", "\n", "Train: LabelList (733 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Valid: LabelList (184 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Test: None, model=Sequential(\n", " (0): Sequential(\n", " (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n", " (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (2): ReLU(inplace=True)\n", " (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n", " (4): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " (downsample): Sequential(\n", " (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " )\n", " (5): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " (downsample): Sequential(\n", " (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (3): Bottleneck(\n", " (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " )\n", " (6): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " (downsample): Sequential(\n", " (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (3): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (4): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (5): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (6): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (7): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (8): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (9): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (10): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (11): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (12): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (13): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (14): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (15): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (16): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (17): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (18): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (19): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (20): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (21): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (22): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " )\n", " (7): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " (downsample): Sequential(\n", " (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace=True)\n", " )\n", " )\n", " )\n", " (1): Sequential(\n", " (0): AdaptiveConcatPool2d(\n", " (ap): AdaptiveAvgPool2d(output_size=1)\n", " (mp): AdaptiveMaxPool2d(output_size=1)\n", " )\n", " (1): Flatten()\n", " (2): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (3): Dropout(p=0.25, inplace=False)\n", " (4): Linear(in_features=4096, out_features=512, bias=True)\n", " (5): ReLU(inplace=True)\n", " (6): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (7): Dropout(p=0.5, inplace=False)\n", " (8): Linear(in_features=512, out_features=7, bias=True)\n", " )\n", "), opt_func=functools.partial(, betas=(0.9, 0.99)), loss_func=FlattenedLoss of CrossEntropyLoss(), metrics=[, Precision(average='macro', pos_label=1, eps=1e-09), Recall(average='macro', pos_label=1, eps=1e-09), FBeta(average='macro', pos_label=1, eps=1e-09, beta=2), KappaScore(weights='quadratic')], true_wd=True, bn_wd=True, wd=0.01, train_bn=True, path=PosixPath('../../../Dataset/Herlev Dataset'), model_dir='models', callback_fns=[functools.partial(, add_time=True, silent=False)], callbacks=[...], layer_groups=[Sequential(\n", " (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n", " (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (2): ReLU(inplace=True)\n", " (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n", " (4): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (5): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (6): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (7): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (8): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (9): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (10): ReLU(inplace=True)\n", " (11): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (12): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (13): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (14): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (15): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (16): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (17): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (18): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (19): ReLU(inplace=True)\n", " (20): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (21): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (22): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (23): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (24): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (25): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (26): ReLU(inplace=True)\n", " (27): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (28): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (29): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (30): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (31): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (32): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (33): ReLU(inplace=True)\n", " (34): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (35): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (36): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (37): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (38): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (39): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (40): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (41): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (42): ReLU(inplace=True)\n", " (43): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (44): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (45): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (46): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (47): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (48): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (49): ReLU(inplace=True)\n", " (50): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (51): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (52): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (53): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (54): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (55): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (56): ReLU(inplace=True)\n", "), Sequential(\n", " (0): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (4): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (5): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (6): ReLU(inplace=True)\n", " (7): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (8): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (9): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (10): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (11): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (12): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (13): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (14): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (15): ReLU(inplace=True)\n", " (16): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (17): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (18): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (19): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (20): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (21): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (22): ReLU(inplace=True)\n", " (23): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (24): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (25): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (26): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (27): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (28): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (29): ReLU(inplace=True)\n", " (30): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (31): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (32): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (33): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (34): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (35): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (36): ReLU(inplace=True)\n", " (37): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (38): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (39): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (40): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (41): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (42): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (43): ReLU(inplace=True)\n", " (44): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (45): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (46): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (47): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (48): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (49): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (50): ReLU(inplace=True)\n", " (51): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (52): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (53): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (54): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (55): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (56): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (57): ReLU(inplace=True)\n", " (58): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (59): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (60): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (61): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (62): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (63): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (64): ReLU(inplace=True)\n", " (65): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (66): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (67): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (68): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (69): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (70): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (71): ReLU(inplace=True)\n", " (72): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (73): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (74): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (75): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (76): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (77): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (78): ReLU(inplace=True)\n", " (79): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (80): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (81): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (82): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (83): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (84): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (85): ReLU(inplace=True)\n", " (86): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (87): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (88): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (89): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (90): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (91): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (92): ReLU(inplace=True)\n", " (93): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (94): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (95): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (96): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (97): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (98): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (99): ReLU(inplace=True)\n", " (100): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (101): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (102): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (103): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (104): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (105): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (106): ReLU(inplace=True)\n", " (107): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (108): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (109): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (110): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (111): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (112): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (113): ReLU(inplace=True)\n", " (114): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (115): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (116): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (117): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (118): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (119): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (120): ReLU(inplace=True)\n", " (121): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (122): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (123): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (124): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (125): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (126): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (127): ReLU(inplace=True)\n", " (128): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (129): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (130): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (131): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (132): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (133): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (134): ReLU(inplace=True)\n", " (135): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (136): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (137): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (138): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (139): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (140): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (141): ReLU(inplace=True)\n", " (142): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (143): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (144): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (145): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (146): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (147): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (148): ReLU(inplace=True)\n", " (149): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (150): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (151): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (152): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (153): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (154): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (155): ReLU(inplace=True)\n", " (156): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (157): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (158): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (159): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (160): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (161): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (162): ReLU(inplace=True)\n", " (163): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (164): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (165): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (166): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (167): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (168): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (169): ReLU(inplace=True)\n", " (170): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (171): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (172): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (173): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (174): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (175): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (176): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (177): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (178): ReLU(inplace=True)\n", " (179): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (180): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (181): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (182): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (183): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (184): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (185): ReLU(inplace=True)\n", "), Sequential(\n", " (0): AdaptiveAvgPool2d(output_size=1)\n", " (1): AdaptiveMaxPool2d(output_size=1)\n", " (2): Flatten()\n", " (3): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (4): Dropout(p=0.25, inplace=False)\n", " (5): Linear(in_features=4096, out_features=512, bias=True)\n", " (6): ReLU(inplace=True)\n", " (7): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (8): Dropout(p=0.5, inplace=False)\n", " (9): Linear(in_features=512, out_features=7, bias=True)\n", ")], add_time=True, silent=False)\n", "loss_scale: 65536\n", "max_noskip: 1000\n", "dynamic: True\n", "clip: None\n", "flat_master: False\n", "max_scale: 16777216\n", "loss_fp32: True], layer_groups=[Sequential(\n", " (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n", " (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (2): ReLU(inplace=True)\n", " (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n", " (4): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (5): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (6): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (7): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (8): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (9): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (10): ReLU(inplace=True)\n", " (11): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (12): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (13): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (14): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (15): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (16): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (17): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (18): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (19): ReLU(inplace=True)\n", " (20): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (21): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (22): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (23): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (24): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (25): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (26): ReLU(inplace=True)\n", " (27): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (28): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (29): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (30): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (31): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (32): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (33): ReLU(inplace=True)\n", " (34): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (35): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (36): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (37): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (38): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (39): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (40): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (41): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (42): ReLU(inplace=True)\n", " (43): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (44): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (45): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (46): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (47): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (48): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (49): ReLU(inplace=True)\n", " (50): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (51): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (52): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (53): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (54): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (55): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (56): ReLU(inplace=True)\n", "), Sequential(\n", " (0): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (4): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (5): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (6): ReLU(inplace=True)\n", " (7): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (8): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (9): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (10): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (11): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (12): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (13): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (14): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (15): ReLU(inplace=True)\n", " (16): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (17): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (18): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (19): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (20): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (21): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (22): ReLU(inplace=True)\n", " (23): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (24): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (25): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (26): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (27): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (28): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (29): ReLU(inplace=True)\n", " (30): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (31): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (32): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (33): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (34): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (35): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (36): ReLU(inplace=True)\n", " (37): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (38): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (39): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (40): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (41): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (42): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (43): ReLU(inplace=True)\n", " (44): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (45): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (46): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (47): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (48): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (49): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (50): ReLU(inplace=True)\n", " (51): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (52): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (53): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (54): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (55): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (56): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (57): ReLU(inplace=True)\n", " (58): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (59): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (60): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (61): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (62): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (63): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (64): ReLU(inplace=True)\n", " (65): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (66): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (67): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (68): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (69): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (70): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (71): ReLU(inplace=True)\n", " (72): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (73): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (74): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (75): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (76): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (77): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (78): ReLU(inplace=True)\n", " (79): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (80): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (81): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (82): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (83): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (84): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (85): ReLU(inplace=True)\n", " (86): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (87): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (88): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (89): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (90): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (91): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (92): ReLU(inplace=True)\n", " (93): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (94): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (95): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (96): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (97): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (98): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (99): ReLU(inplace=True)\n", " (100): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (101): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (102): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (103): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (104): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (105): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (106): ReLU(inplace=True)\n", " (107): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (108): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (109): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (110): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (111): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (112): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (113): ReLU(inplace=True)\n", " (114): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (115): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (116): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (117): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (118): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (119): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (120): ReLU(inplace=True)\n", " (121): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (122): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (123): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (124): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (125): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (126): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (127): ReLU(inplace=True)\n", " (128): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (129): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (130): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (131): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (132): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (133): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (134): ReLU(inplace=True)\n", " (135): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (136): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (137): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (138): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (139): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (140): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (141): ReLU(inplace=True)\n", " (142): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (143): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (144): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (145): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (146): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (147): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (148): ReLU(inplace=True)\n", " (149): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (150): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (151): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (152): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (153): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (154): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (155): ReLU(inplace=True)\n", " (156): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (157): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (158): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (159): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (160): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (161): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (162): ReLU(inplace=True)\n", " (163): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (164): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (165): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (166): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (167): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (168): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (169): ReLU(inplace=True)\n", " (170): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (171): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (172): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (173): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (174): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (175): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (176): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (177): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (178): ReLU(inplace=True)\n", " (179): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (180): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (181): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (182): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (183): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (184): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (185): ReLU(inplace=True)\n", "), Sequential(\n", " (0): AdaptiveAvgPool2d(output_size=1)\n", " (1): AdaptiveMaxPool2d(output_size=1)\n", " (2): Flatten()\n", " (3): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (4): Dropout(p=0.25, inplace=False)\n", " (5): Linear(in_features=4096, out_features=512, bias=True)\n", " (6): ReLU(inplace=True)\n", " (7): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (8): Dropout(p=0.5, inplace=False)\n", " (9): Linear(in_features=512, out_features=7, bias=True)\n", ")], add_time=True, silent=False)" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "learner = cnn_learner(fold_data, models.resnet101, metrics=our_metrics).to_fp16()\n", "learner" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 80.00% [8/10 00:19<00:04]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
03.175865#na#00:02
13.185722#na#00:02
23.192070#na#00:02
33.041921#na#00:02
42.768011#na#00:02
52.565398#na#00:02
62.415970#na#00:02
76.264168#na#00:02

\n", "\n", "

\n", " \n", " \n", " 9.09% [1/11 00:01<00:11 8.3028]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
03.0172152.0018650.2717390.3042050.2757740.2737160.22326300:03
12.8828521.7877800.3750000.3989810.4290000.4163280.47430400:03
22.6982051.7864690.3913040.3848380.4347070.4192870.47557500:03
32.4418421.8886090.4510870.4504680.4911100.4752800.51938900:03
42.2796991.7729310.4673910.4956210.4972360.4897240.58053200:03
52.1064911.9240600.5163040.5313820.5474340.5428970.60729100:03
61.9885612.0081940.4728260.4945280.5009740.4982680.55815500:03
71.8820871.8641840.4836960.5010550.5144600.5080720.60251200:03
81.8031282.0789470.4673910.4877240.4870170.4853750.55571800:03
91.7142111.7764860.5108700.5223290.5277920.5258340.57342100:03
101.6462041.8690660.5380430.5515840.5499080.5470830.66712200:03
111.6003391.6057270.4891300.5166000.5189260.5179850.63600300:03
121.5601331.4988500.5108700.5495960.5226940.5259600.59888900:03
131.5095201.3928920.5434780.5462140.5535480.5461850.67590500:03
141.4506241.4417190.5543480.5688050.5618130.5579250.70003500:03
151.3785671.4479370.5108700.5271890.5349990.5299850.61854800:03
161.3413911.7778210.5271740.5432520.5458640.5396240.62928100:03
171.3046491.6076940.5326090.5679800.5436020.5451840.63828600:03
181.2476951.4980670.5434780.5847540.5702310.5719320.63694200:03
191.2269751.5367720.5163040.5572920.5353160.5372350.69375300:03
201.2106101.4586500.5652170.5856030.5781950.5762420.70633200:03
211.1734961.6075800.5706520.5966570.5845380.5830740.66024200:03
221.1272621.4992730.5543480.5903670.5755340.5757380.62289500:03
231.0770051.3676710.5978260.6369720.6162680.6169600.67845900:03
241.0388551.4595300.5869570.6353120.6120260.6137410.67061100:03
251.0132331.4879000.5869570.6326450.6142090.6159140.68802000:03
260.9908121.5751910.5706520.6090160.5862160.5873480.69129300:03
270.9661461.4512260.5706520.6150080.5876960.5893360.68709300:03
280.9487841.4323860.5652170.6121850.5880180.5906730.69236300:03
290.9356461.6928010.5815220.6127320.6012050.6020290.70844300:03
300.9109781.6637530.5923910.6248290.6096300.6106470.70489500:03
310.9027441.6596660.5978260.6384390.6134910.6156130.72320800:03
320.8920071.5823510.5923910.6325460.6100260.6119340.72369000:03
330.8588161.3690610.5923910.6372470.6133080.6164050.73147000:03
340.8439081.3810450.5978260.6404760.6169710.6199520.72663800:03
350.8426451.3554650.6086960.6511510.6255940.6286610.71698000:03
360.8390331.3258460.6032610.6445160.6239160.6265650.74563900:03
370.8251501.3116030.5923910.6301920.6161940.6180580.70329800:03
380.8246401.2982440.5923910.6267070.6152930.6167270.71983800:03
390.8354351.4680440.5815220.6085160.6019080.6014690.70381800:03
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.27173912525177.\n", "Better model found at epoch 1 with accuracy value: 0.375.\n", "Better model found at epoch 2 with accuracy value: 0.3913043439388275.\n", "Better model found at epoch 3 with accuracy value: 0.45108696818351746.\n", "Better model found at epoch 4 with accuracy value: 0.46739131212234497.\n", "Better model found at epoch 5 with accuracy value: 0.5163043737411499.\n", "Better model found at epoch 10 with accuracy value: 0.5380434989929199.\n", "Better model found at epoch 13 with accuracy value: 0.54347825050354.\n", "Better model found at epoch 14 with accuracy value: 0.554347813129425.\n", "Better model found at epoch 20 with accuracy value: 0.5652173757553101.\n", "Better model found at epoch 21 with accuracy value: 0.570652186870575.\n", "Better model found at epoch 23 with accuracy value: 0.5978260636329651.\n", "Better model found at epoch 35 with accuracy value: 0.6086956262588501.\n" ] } ], "source": [ "learner.fit_one_cycle(40, max_lr=slice(1e-03), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold1-stage1\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold1-stage1\")" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 70.00% [7/10 00:18<00:08]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.839226#na#00:02
10.804976#na#00:02
20.791369#na#00:02
30.767239#na#00:02
40.835104#na#00:02
51.117921#na#00:02
61.742246#na#00:02

\n", "\n", "

\n", " \n", " \n", " 54.55% [6/11 00:02<00:01 2.9747]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold1-stage1\")\n", "learner = to_fp16(learner)\n", "learner.unfreeze()\n", "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.8111111.4279680.5815220.6130650.6021060.6029450.71201400:03
10.8153181.3816560.5815220.6220160.6019080.6038220.73561200:03
20.8142511.3702010.5923910.6275880.6129120.6145850.71374300:03
30.8226441.2965090.5978260.6388010.6169710.6196980.73497500:03
40.7942621.2244400.6086960.6406480.6292570.6298700.73070700:04
50.7737731.1877360.6304350.6706440.6463040.6488930.72544100:03
60.7611211.1311220.6195650.6599780.6387800.6410860.75085200:03
70.7371141.1088400.6250000.6633360.6437260.6446210.72346300:03
80.7359371.1996220.6141300.6546220.6370800.6377780.71771600:03
90.7110521.3122730.5923910.6347340.6141940.6162320.70186800:03
100.6823751.0669410.6086960.6503300.6369830.6379200.73547200:03
110.6731181.0203500.6521740.6947270.6676960.6712860.76579800:03
120.6563481.0299070.6250000.6682070.6373820.6416290.73528600:03
130.6420471.0482180.6304350.6766270.6486850.6521410.74470900:04
140.6136381.0333920.6195650.6759010.6496870.6529040.72911800:03
150.5834751.0094710.6467390.7026660.6728800.6763470.75703300:03
160.5579061.0469410.6304350.6747240.6502670.6525110.74982600:04
170.5373571.0363990.6630430.6977290.6826990.6838640.77556300:03
180.5185361.0157120.6793480.7154020.7039070.7048060.78414700:03
190.5003470.9866580.6684780.7122970.6931010.6957900.75744500:04
200.4916910.9898580.6467390.6945050.6725730.6754620.77253200:03
210.4681670.9867740.6793480.7225850.7049680.7066180.77773200:03
220.4503050.9734820.6902170.7290380.7105300.7113630.78604700:03
230.4328540.9566860.6902170.7314710.7161700.7168870.78927600:03
240.4129180.9714420.6793480.7226910.7086460.7094260.78438300:03
250.4027711.0053470.6956520.7327780.7244120.7241600.79485700:03
260.3914611.0167270.6902170.7301700.7214520.7216950.80188500:04
270.3720341.0137370.6847830.7259090.7112480.7129050.80239100:03
280.3606670.9960760.6847830.7250720.7101490.7115420.80245200:03
290.3468611.0012680.6902170.7282270.7155910.7171430.81887000:03
300.3375901.0295670.6847830.7241690.7044860.7069770.81253100:03
310.3294671.0172080.7173910.7515930.7377820.7391480.82178600:03
320.3129011.0086870.7065220.7392160.7206330.7227810.81971300:03
330.3009021.0142190.7119570.7511350.7297380.7312710.81514300:03
340.2990681.0256930.7010870.7418470.7124140.7163650.79799200:04
350.2893861.0318710.6956520.7334430.7067660.7101060.79638400:04
360.2831581.0027420.7173910.7549180.7345000.7361830.81313400:03
370.2834421.0175450.7228260.7567950.7381630.7400260.83322700:03
380.2757741.0262710.7228260.7595440.7381630.7399120.81998300:03
390.2795101.0201820.7065220.7493790.7267780.7288100.81349500:04
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.58152174949646.\n", "Better model found at epoch 2 with accuracy value: 0.592391312122345.\n", "Better model found at epoch 3 with accuracy value: 0.5978260636329651.\n", "Better model found at epoch 4 with accuracy value: 0.6086956262588501.\n", "Better model found at epoch 5 with accuracy value: 0.6304348111152649.\n", "Better model found at epoch 11 with accuracy value: 0.6521739363670349.\n", "Better model found at epoch 17 with accuracy value: 0.6630434989929199.\n", "Better model found at epoch 18 with accuracy value: 0.679347813129425.\n", "Better model found at epoch 22 with accuracy value: 0.6902173757553101.\n", "Better model found at epoch 25 with accuracy value: 0.695652186870575.\n", "Better model found at epoch 31 with accuracy value: 0.717391312122345.\n", "Better model found at epoch 37 with accuracy value: 0.7228260636329651.\n" ] } ], "source": [ "learner.fit_one_cycle(40, max_lr=slice(2e-05, 1e-04), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold1-stage2\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold1-stage2\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Fold-2" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "ImageDataBunch;\n", "\n", "Train: LabelList (733 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Valid: LabelList (184 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Test: None" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "fold_idxs = idxs[1]\n", "fold_data = (ImageList.from_folder(data_path)\n", " .split_by_idxs(fold_idxs[0], fold_idxs[1])\n", " .label_from_folder()\n", " .transform(tfms, size=64)\n", " .databunch(bs=64)\n", " .normalize(imagenet_stats))\n", "fold_data" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 70.00% [7/10 00:16<00:06]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.552091#na#00:02
10.583279#na#00:02
20.563949#na#00:02
30.550634#na#00:02
40.542706#na#00:02
50.634392#na#00:02
61.349554#na#00:02

\n", "\n", "

\n", " \n", " \n", " 0.00% [0/11 00:00<00:00]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold1-stage2\")\n", "learner.data = fold_data\n", "learner.freeze()\n", "learner = to_fp16(learner)\n", "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.5442550.0992970.9673910.9746640.9732790.9733310.97948300:03
10.5295770.0964210.9673910.9751560.9697990.9705430.98188700:03
20.5336320.0960030.9673910.9751560.9697990.9705430.98188700:03
30.5199770.0992400.9673910.9738260.9732790.9731540.98191700:03
40.5172490.0963910.9728260.9786270.9769420.9771130.98274000:03
50.5223440.1046640.9673910.9738260.9732790.9731540.98191700:03
60.5408900.1032950.9673910.9738260.9732790.9731540.98191700:03
70.5285550.1035590.9673910.9738260.9732790.9731540.98191700:03
80.5317300.1009860.9673910.9738260.9732790.9731540.98191700:03
90.5341980.0989630.9673910.9738260.9732790.9731540.98191700:03
100.5201460.1062450.9565220.9652520.9657550.9651500.97378000:03
110.5226840.1010780.9673910.9738260.9732790.9731540.98191700:03
120.5110770.1066980.9565220.9652520.9657550.9651500.97378000:03
130.5039480.1056530.9619570.9695530.9696160.9691870.97463900:03
140.5011030.1137770.9565220.9655060.9659530.9652000.96742900:03
150.5042960.1214360.9565220.9655060.9659530.9652000.96742900:03
160.4925360.1238670.9565220.9657330.9657550.9651720.96738100:03
170.4783090.1189260.9619570.9695530.9696160.9691870.97463900:03
180.4732350.1230410.9565220.9654650.9646900.9644380.97386200:03
190.4661110.1245790.9565220.9655670.9659530.9653410.97140900:03
200.4601870.1296840.9456520.9571040.9584290.9572970.96335500:03
210.4531370.1256880.9510870.9612050.9620920.9611630.96656300:03
220.4464920.1248220.9565220.9652520.9657550.9651500.97378000:03
230.4461470.1270110.9510870.9611640.9620920.9613170.97054600:03
240.4481290.1303570.9510870.9611640.9620920.9613170.97054600:03
250.4514480.1310580.9565220.9655060.9659530.9652000.96742900:03
260.4334460.1325480.9510870.9615070.9622900.9613220.96422500:03
270.4423280.1349490.9456520.9571040.9584290.9572970.96335500:03
280.4400950.1354740.9456520.9571040.9584290.9572970.96335500:03
290.4310080.1383080.9456520.9571040.9584290.9572970.96335500:03
300.4274550.1415150.9456520.9571040.9584290.9572970.96335500:03
310.4216680.1420640.9456520.9571040.9584290.9572970.96335500:03
320.4106570.1427370.9456520.9571040.9584290.9572970.96335500:03
330.4091940.1491210.9456520.9571040.9584290.9572970.96335500:03
340.4038770.1450310.9347830.9488890.9511030.9493120.95934100:03
350.4046290.1451840.9456520.9576850.9586270.9574200.96103100:03
360.4114690.1392830.9402170.9527530.9536670.9529080.95582000:03
370.4115270.1477810.9456520.9571040.9584290.9572970.96335500:03
380.4065110.1411760.9402170.9531920.9547660.9534070.96015700:03
390.4079700.1421540.9402170.9531920.9547660.9534070.96015700:03
400.4121560.1418670.9456520.9572660.9584290.9574590.96732300:03
410.4054860.1449290.9456520.9571040.9584290.9572970.96335500:03
420.4065410.1410690.9402170.9531920.9547660.9534070.96015700:03
430.4070090.1364550.9456520.9571040.9584290.9572970.96335500:03
440.4107280.1420340.9456520.9571040.9584290.9572970.96335500:03
450.4118370.1471460.9456520.9571040.9584290.9572970.96335500:03
460.4178740.1454960.9402170.9531920.9547660.9534070.96015700:03
470.4113790.1402110.9456520.9576850.9586270.9574200.96103100:03
480.4034430.1384740.9456520.9576850.9586270.9574200.96103100:03
490.4019790.1428490.9510870.9617600.9622900.9614720.96818900:03
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.967391312122345.\n", "Better model found at epoch 4 with accuracy value: 0.9728260636329651.\n" ] } ], "source": [ "learner.fit_one_cycle(50, max_lr=slice(1e-04), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold2-stage1\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold2-stage1\")" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 70.00% [7/10 00:19<00:08]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.527440#na#00:02
10.540798#na#00:02
20.514713#na#00:02
30.520216#na#00:03
40.644984#na#00:02
50.962250#na#00:02
61.499816#na#00:02

\n", "\n", "

\n", " \n", " \n", " 18.18% [2/11 00:01<00:05 1.6560]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold2-stage1\")\n", "learner = to_fp16(learner)\n", "learner.unfreeze()\n", "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.5319850.0992210.9673910.9751560.9697990.9705430.98188700:03
10.5021150.0979790.9673910.9738260.9732790.9731540.98191700:03
20.5161780.0977080.9619570.9702690.9661360.9665960.98106300:03
30.5149620.1024330.9510870.9618250.9588100.9588090.97054200:03
40.5064590.1042380.9565220.9672360.9624730.9629800.97537500:03
50.4959150.1145950.9456520.9579180.9551470.9549520.96731800:03
60.4781540.1232780.9456520.9576650.9551470.9548030.96335000:03
70.4580300.1373110.9510870.9614190.9611910.9609210.96388800:03
80.4455100.1399000.9456520.9575120.9575280.9570640.96066400:03
90.4175710.1464890.9510870.9624500.9611910.9611060.96148600:03
100.4098650.1678740.9347830.9497390.9467410.9472070.94788500:04
110.3877690.1647170.9239130.9413340.9370340.9373590.95058700:03
120.3706200.1744020.9184780.9369120.9331730.9333920.94968400:03
130.3466210.1793480.9184780.9334990.9344550.9341030.93029800:03
140.3366190.2107760.9184780.9344280.9335540.9334880.92341700:03
150.3197350.2508270.9239130.9395590.9383160.9381810.93516900:03
160.3255180.3017620.9021740.9153420.9207230.9184710.90337500:03
170.3176210.2946580.8967390.9108960.9138010.9123850.94447200:03
180.3118040.2796590.9021740.9197390.9220390.9210210.92648600:03
190.2898790.2500470.9130430.9328900.9269840.9274620.93474300:03
200.2705690.2436500.9076090.9289790.9212680.9220530.92894100:03
210.2610950.2990690.9130430.9294250.9297080.9289850.94664100:03
220.2514210.3105810.9076090.9273200.9122350.9140080.95083700:03
230.2452780.3118270.8967390.9160000.9101230.9106070.94640800:03
240.2379710.3477970.8750000.8993980.8998670.8995490.92011000:04
250.2330290.3637870.8913040.9122930.9127250.9119870.94797800:03
260.2265860.3465050.8967390.9195970.9165530.9167920.93956400:03
270.2177490.3208580.9130430.9314970.9315290.9310540.94664900:03
280.2016100.3612600.9130430.9320700.9306280.9305950.93362500:03
290.1946660.3452360.8913040.9147320.9132520.9128370.92833400:03
300.1899890.3169830.9184780.9376490.9353900.9353140.94746600:03
310.1853340.3045390.9184780.9368100.9376070.9367540.94535800:03
320.1663080.3302390.9021740.9260200.9183950.9190340.94280600:03
330.1548840.3561680.8804350.9043870.8933260.8945120.93323000:03
340.1506880.3566380.8913040.9133300.9008840.9016050.94433100:03
350.1426140.3592870.8967390.9172270.9068750.9077710.95105700:03
360.1350300.3686870.8967390.9157970.9127360.9124270.93821600:03
370.1347710.3717500.8913040.9121900.9078100.9076260.93750000:03
380.1289430.3757160.8858700.9068200.9052120.9044730.93648700:03
390.1287950.3647220.9130430.9327920.9326280.9317460.94706100:03
400.1242830.3759440.9076090.9245250.9278660.9264180.92997200:03
410.1217360.3692670.8967390.9164610.9168620.9161520.93197700:03
420.1201510.3438480.8967390.9195780.9189150.9179150.93858900:03
430.1204770.3445430.9076090.9284430.9277020.9268180.94244700:03
440.1106230.3384230.9076090.9245250.9278660.9264180.92997200:03
450.1085710.3330690.8967390.9195780.9189150.9179150.93858900:03
460.1049470.3449900.9021740.9236760.9238410.9226960.93930300:03
470.1038840.3458290.9021740.9236760.9238410.9226960.93930300:03
480.0972130.3486060.8967390.9195780.9189150.9179150.93858900:03
490.1006650.3408460.8967390.9195780.9189150.9179150.93858900:03
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.967391312122345.\n" ] } ], "source": [ "learner.fit_one_cycle(50, max_lr=slice(1e-05, 2.5e-04), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold2-stage2\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold2-stage2\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Fold-3" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "ImageDataBunch;\n", "\n", "Train: LabelList (734 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Valid: LabelList (183 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Test: None" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "fold_idxs = idxs[2]\n", "fold_data = (ImageList.from_folder(data_path)\n", " .split_by_idxs(fold_idxs[0], fold_idxs[1])\n", " .label_from_folder()\n", " .transform(tfms, size=64)\n", " .databunch(bs=64)\n", " .normalize(imagenet_stats))\n", "fold_data" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 70.00% [7/10 00:15<00:06]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.509085#na#00:02
10.538736#na#00:02
20.532676#na#00:02
30.545773#na#00:02
40.528709#na#00:02
50.611814#na#00:02
61.181265#na#00:02

\n", "\n", "

\n", " \n", " \n", " 45.45% [5/11 00:01<00:01 1.7955]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold2-stage2\")\n", "learner.data = fold_data\n", "learner.freeze()\n", "learner = to_fp16(learner)\n", "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.5504930.0600400.9781420.9819990.9849210.9839660.99000600:03
10.5500400.0631200.9781420.9819000.9839630.9834430.99003800:03
20.5468850.0656530.9726780.9776790.9813490.9801020.98259400:03
30.5196080.0659800.9781420.9819990.9849210.9839660.99000600:03
40.5159950.0732750.9781420.9819990.9849210.9839660.99000600:03
50.5103970.0884540.9726780.9748560.9813490.9795480.98919200:03
60.4963630.0989000.9617490.9658830.9714970.9700690.98510800:03
70.4863280.1112600.9508200.9517850.9570470.9555580.98339600:03
80.4857720.1120860.9672130.9731230.9750680.9744760.98591100:03
90.4732880.1087190.9672130.9703460.9768200.9749860.97944600:03
100.4815220.1152960.9672130.9676540.9756300.9736130.97348500:03
110.4758810.1357220.9562840.9591220.9619730.9608170.97765200:03
120.4669210.1168690.9672130.9679420.9768200.9744500.98598100:03
130.4585110.1618920.9562840.9525430.9637250.9602980.98428000:03
140.4476310.1370740.9672130.9701040.9768200.9749750.98595600:03
150.4372990.1510690.9453550.9492200.9609470.9575100.96124700:03
160.4320580.1660820.9398910.9469800.9470760.9457570.96053400:03
170.4305050.1677710.9398910.9497340.9558570.9535160.96259400:03
180.4254340.1773000.9398910.9481290.9581690.9543140.96248200:03
190.4168280.1598760.9398910.9500650.9523060.9509330.96258900:03
200.4155000.1574660.9398910.9405070.9550630.9512030.96723700:03
210.4101360.1724240.9398910.9437350.9505790.9473900.97941500:03
220.4062210.1707260.9562840.9574530.9567850.9561300.96957600:03
230.3947830.2307470.9234970.9262560.9351230.9312470.94563800:03
240.3879270.1884910.9453550.9504740.9564390.9543520.96142200:03
250.3780540.1839060.9508200.9523190.9604070.9580440.96876400:03
260.3738680.1836300.9344260.9387860.9488990.9453270.95330300:03
270.3700240.1653300.9508200.9529330.9608040.9582840.96242500:03
280.3581610.1533290.9508200.9550830.9617610.9594860.96475800:03
290.3517400.1578100.9508200.9577680.9558090.9554010.94452100:03
300.3456230.1511650.9508200.9563430.9612000.9590720.95859200:03
310.3388390.1579580.9453550.9506540.9568350.9548710.95738500:03
320.3326280.1612830.9398910.9463910.9520730.9504640.94977300:03
330.3300360.1820450.9289620.9351850.9453270.9414080.94618400:03
340.3293190.1883910.9344260.9439980.9450950.9431440.92958000:03
350.3352950.1976060.9289620.9355960.9339330.9325510.93500600:03
360.3298430.1571540.9398910.9451300.9536610.9506830.95691000:03
370.3108610.1529920.9453550.9498110.9562740.9537800.95535700:03
380.3144600.1613520.9344260.9430840.9492960.9470640.95343700:03
390.3116560.1577690.9398910.9474720.9532640.9515090.95656800:03
400.2996220.1661170.9453550.9488510.9572320.9546210.95752400:03
410.2991050.1669150.9398910.9444180.9532640.9504810.95664000:03
420.2967460.1673680.9344260.9406960.9496920.9465440.94951400:03
430.3001580.1689970.9344260.9405630.9492960.9465810.95329700:03
440.3001370.1685080.9398910.9444180.9532640.9504810.95664000:03
450.2965640.1691650.9398910.9444180.9532640.9504810.95664000:03
460.2922100.1720170.9344260.9400700.9483380.9458310.95582400:03
470.2981410.1665470.9453550.9488510.9572320.9546210.95752400:03
480.2996400.1605150.9398910.9444180.9532640.9504810.95664000:03
490.2954600.1608310.9289620.9364240.9443700.9419370.95248100:03
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.9781420826911926.\n" ] } ], "source": [ "learner.fit_one_cycle(50, max_lr=slice(1e-03), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold3-stage1\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold3-stage1\")" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 60.00% [6/10 00:15<00:10]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.589510#na#00:02
10.567367#na#00:02
20.573371#na#00:02
30.568437#na#00:02
40.745412#na#00:02
51.025183#na#00:02

\n", "\n", "

\n", " \n", " \n", " 90.91% [10/11 00:02<00:00 1.7860]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold3-stage1\")\n", "learner = to_fp16(learner)\n", "learner.unfreeze()\n", "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.4884840.0602330.9726780.9775790.9803910.9795790.98264900:03
10.5183790.0663210.9726780.9775790.9803910.9795790.98264900:03
20.5247080.0729200.9672130.9679610.9704710.9695620.98830000:03
30.5188340.0623450.9726780.9776790.9813490.9801020.98259400:03
40.5279880.0646290.9726780.9776790.9813490.9801020.98259400:03
50.5371930.0652860.9781420.9819990.9849210.9839660.99000600:03
60.5425050.0625770.9836070.9861750.9888890.9881460.99085300:03
70.5339860.0703160.9726780.9748560.9813490.9795480.98919200:03
80.5171610.0695320.9726780.9748560.9813490.9795480.98919200:03
90.4920700.0742730.9672130.9705360.9777780.9756540.98179700:03
100.4794950.0741350.9617490.9660220.9728520.9709850.98099900:03
110.4694530.0802840.9617490.9636410.9668990.9656990.98087000:03
120.4587850.0761540.9781420.9819990.9849210.9839660.99000600:03
130.4566450.0881310.9726780.9724750.9753970.9742310.98911800:03
140.4439070.0790360.9781420.9819990.9849210.9839660.99000600:03
150.4331650.0878800.9726780.9724750.9753970.9742310.98911800:03
160.4185490.0861290.9672130.9705360.9777780.9756540.98179700:03
170.4049460.0877200.9617490.9660220.9728520.9709850.98099900:03
180.3984250.0843670.9726780.9774850.9799950.9792960.98919000:03
190.3841360.1003850.9617490.9587700.9672960.9648730.98752700:03
200.3706350.1103200.9453550.9452270.9534760.9510640.98262300:03
210.3634790.1185450.9508200.9516900.9570470.9554810.98342200:03
220.3510030.1489770.9562840.9560370.9619730.9602500.98422600:03
230.3405850.1245440.9672130.9679610.9704710.9695620.98830000:03
240.3316160.0887500.9672130.9703420.9764230.9748780.98837900:03
250.3232670.0944770.9562840.9543560.9633280.9607270.98667500:03
260.3137230.0961080.9617490.9638800.9728520.9704610.98757000:03
270.3055290.1045020.9726780.9699380.9748360.9734140.98918900:03
280.2979730.1088380.9672130.9652330.9708680.9692910.98833900:03
290.2944570.1083600.9617490.9605840.9659410.9643760.98507900:03
300.2873040.1177350.9617490.9605840.9659410.9643760.98507900:03
310.2886680.0942770.9562840.9610100.9665710.9650480.98191400:03
320.2796220.1297870.9562840.9561740.9610150.9594180.98185200:03
330.2754490.1323680.9617490.9587820.9663380.9639290.98270600:03
340.2642290.1039690.9562840.9525620.9633280.9600620.98417900:03
350.2629860.1341700.9617490.9580900.9659410.9636940.98259600:03
360.2597610.1122540.9562840.9541170.9633280.9605560.98422800:03
370.2447980.1109300.9617490.9605800.9668990.9649730.98503200:03
380.2454440.1001670.9508200.9545480.9629990.9606300.98112400:03
390.2368640.1045430.9508200.9545480.9629990.9606300.98112400:03
400.2364330.1499000.9398910.9391490.9499040.9464790.97282800:03
410.2255520.1442020.9562840.9536560.9619730.9595550.98173900:03
420.2254510.1228070.9617490.9605800.9668990.9649730.98503200:03
430.2303840.1328000.9562840.9541220.9623700.9599580.98427800:03
440.2265400.1209590.9562840.9541220.9623700.9599580.98427800:03
450.2238860.1393170.9562840.9522150.9623700.9593070.98179900:03
460.2177870.0952490.9617490.9636410.9728520.9702900.98513300:03
470.2191960.1129280.9617490.9605800.9668990.9649730.98503200:03
480.2250480.1143690.9562840.9560610.9619730.9600910.98179400:03
490.2256150.1315170.9617490.9605800.9668990.9649730.98503200:03
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.9726775884628296.\n", "Better model found at epoch 5 with accuracy value: 0.9781420826911926.\n", "Better model found at epoch 6 with accuracy value: 0.9836065769195557.\n" ] } ], "source": [ "learner.fit_one_cycle(50, max_lr=slice(2e-06, 1e-04), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold3-stage2\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold3-stage2\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Fold-4" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "ImageDataBunch;\n", "\n", "Train: LabelList (734 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Valid: LabelList (183 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Test: None" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "fold_idxs = idxs[3]\n", "fold_data = (ImageList.from_folder(data_path)\n", " .split_by_idxs(fold_idxs[0], fold_idxs[1])\n", " .label_from_folder()\n", " .transform(tfms, size=64)\n", " .databunch(bs=64)\n", " .normalize(imagenet_stats))\n", "fold_data" ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 70.00% [7/10 00:15<00:06]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.457939#na#00:02
10.473938#na#00:02
20.473695#na#00:02
30.467325#na#00:02
40.475849#na#00:02
50.582796#na#00:02
61.102351#na#00:02

\n", "\n", "

\n", " \n", " \n", " 36.36% [4/11 00:01<00:02 1.6080]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold3-stage2\")\n", "learner.data = fold_data\n", "learner.freeze()\n", "learner = to_fp16(learner)\n", "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.4818760.0591060.9890710.9889960.9901480.9897310.99584400:03
10.5098000.0536490.9945360.9961390.9950740.9952420.99916700:03
20.5047980.0538160.9945360.9961390.9950740.9952420.99916700:03
30.5049920.0559180.9945360.9961390.9950740.9952420.99916700:03
40.4953170.0551801.0000001.0000001.0000001.0000001.00000000:03
50.4861790.0560960.9945360.9961390.9950740.9952420.99916700:03
60.4765100.0554081.0000001.0000001.0000001.0000001.00000000:03
70.4742600.0580850.9945360.9961390.9950740.9952420.99916700:03
80.4748660.0567720.9945360.9961390.9950740.9952420.99916700:03
90.4673050.0590040.9890710.9889960.9901480.9897310.99584400:03
100.4651600.0565960.9945360.9961390.9950740.9952420.99916700:03
110.4639820.0627280.9945360.9928570.9950740.9945440.99667000:03
120.4661120.0618450.9890710.9889960.9901480.9897310.99584400:03
130.4676850.0600810.9945360.9961390.9950740.9952420.99916700:03
140.4550180.0607330.9945360.9961390.9950740.9952420.99916700:03
150.4536100.0680120.9890710.9889960.9901480.9897310.99584400:03
160.4417250.0665031.0000001.0000001.0000001.0000001.00000000:03
170.4305800.0725530.9945360.9904760.9924810.9919090.99916900:03
180.4254170.0821080.9890710.9866150.9875550.9871510.99834000:03
190.4265160.0903280.9836070.9820070.9839840.9833340.99091400:03
200.4326310.1346300.9836070.9820070.9839840.9833340.99091400:03
210.4306050.1363810.9836070.9820070.9839840.9833340.99091400:03
220.4223050.1215300.9890710.9858680.9889100.9880910.99172700:03
230.4197510.1337530.9890710.9858680.9889100.9880910.99172700:03
240.4100740.1350410.9890710.9858680.9889100.9880910.99172700:03
250.4077120.1178650.9890710.9858680.9889100.9880910.99172700:03
260.4147920.1232450.9890710.9858680.9889100.9880910.99172700:03
270.4030840.1077930.9781420.9744880.9790580.9778700.98763200:03
280.4000860.0820760.9836070.9783490.9839840.9826830.98843800:03
290.4039370.0710640.9945360.9953920.9964290.9961820.99252900:03
300.4052430.0682420.9945360.9953920.9964290.9961820.99252900:03
310.4052310.0728060.9890710.9882490.9915020.9907260.98922800:03
320.4057480.0755240.9836070.9822960.9839840.9834230.99090000:03
330.3962320.1080780.9836070.9783490.9839840.9826830.98843800:03
340.3954950.1014080.9836070.9783490.9839840.9826830.98843800:03
350.3913570.1043520.9836070.9783490.9839840.9826830.98843800:03
360.3903750.0804170.9836070.9783490.9839840.9826830.98843800:03
370.3855980.0854630.9836070.9843880.9865760.9859130.98841800:03
380.3794940.0804520.9836070.9843880.9865760.9859130.98841800:03
390.3753040.0805180.9836070.9843880.9865760.9859130.98841800:03
400.3801250.0790420.9836070.9843880.9865760.9859130.98841800:03
410.3791240.0713000.9890710.9882490.9915020.9907260.98922800:03
420.3792030.0741620.9945360.9953920.9964290.9961820.99252900:03
430.3793700.0756260.9890710.9915310.9915020.9914250.99171200:03
440.3789900.0716060.9945360.9953920.9964290.9961820.99252900:03
450.3913360.0735400.9890710.9915310.9915020.9914250.99171200:03
460.3895090.0838660.9781420.9744880.9790580.9778700.98763200:03
470.3962970.0829340.9836070.9783490.9839840.9826830.98843800:03
480.3958070.0742640.9890710.9915310.9915020.9914250.99171200:03
490.3936080.0700430.9945360.9953920.9964290.9961820.99252900:03
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.9890710115432739.\n", "Better model found at epoch 1 with accuracy value: 0.994535505771637.\n", "Better model found at epoch 4 with accuracy value: 1.0.\n" ] } ], "source": [ "learner.fit_one_cycle(50, max_lr=slice(1e-04), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold4-stage1\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold4-stage1\")" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 70.00% [7/10 00:18<00:07]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.521343#na#00:02
10.479779#na#00:02
20.469711#na#00:02
30.453941#na#00:02
40.625390#na#00:02
51.002249#na#00:02
61.532500#na#00:02

\n", "\n", "

\n", " \n", " \n", " 27.27% [3/11 00:01<00:03 1.8000]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold4-stage1\")\n", "learner = to_fp16(learner)\n", "learner.unfreeze()\n", "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.4220690.0553870.9945360.9961390.9950740.9952420.99916700:03
10.4419910.0549250.9945360.9961390.9950740.9952420.99916700:03
20.4557810.0552550.9945360.9961390.9950740.9952420.99916700:03
30.4449710.0546471.0000001.0000001.0000001.0000001.00000000:03
40.4489710.0543321.0000001.0000001.0000001.0000001.00000000:03
50.4395530.0539500.9945360.9961390.9950740.9952420.99916700:03
60.4346760.0565980.9945360.9961390.9950740.9952420.99916700:03
70.4295300.0589280.9945360.9961390.9950740.9952420.99916700:03
80.4204310.0569061.0000001.0000001.0000001.0000001.00000000:03
90.4283690.0593741.0000001.0000001.0000001.0000001.00000000:03
100.4165770.0647570.9890710.9889960.9901480.9897310.99584400:03
110.3973150.0681770.9945360.9961390.9950740.9952420.99916700:04
120.3871020.0727730.9836070.9875550.9879310.9877320.99501100:03
130.3799140.0812380.9836070.9855120.9852220.9849400.99501300:03
140.3711460.0810920.9945360.9961390.9950740.9952420.99916700:03
150.3614910.0772010.9836070.9879590.9865760.9867010.99088400:03
160.3515970.0798170.9836070.9878730.9865760.9866210.99089800:03
170.3319810.0860270.9836070.9879590.9839840.9846460.99086700:03
180.3156720.0912990.9836070.9851360.9826290.9829890.99500400:03
190.2990120.1184470.9672130.9715520.9679670.9682790.96710700:03
200.2926570.2520140.9508200.9616750.9492420.9503100.96463300:03
210.2820830.2708630.9508200.9599810.9567400.9565650.97044500:03
220.2690540.1297770.9508200.9596110.9594960.9588530.97031100:03
230.2602200.1052020.9508200.9611490.9598930.9591330.96388700:03
240.2545310.1059310.9508200.9547200.9573010.9560590.96395100:03
250.2431520.1301060.9453550.9474360.9464220.9453180.96939100:03
260.2374510.1142490.9453550.9475210.9525390.9508220.95069400:03
270.2223420.0824330.9726780.9764980.9782430.9777010.97348500:03
280.2159080.0719750.9890710.9910710.9928570.9923480.98512900:03
290.2101830.1009000.9672130.9649590.9715180.9699210.97945200:03
300.2035570.1038180.9726780.9730440.9750890.9744270.98926200:03
310.1902040.1270490.9672130.9665100.9705600.9691750.97951600:03
320.1840940.1217030.9672130.9715820.9692050.9687750.98849100:03
330.1726080.1050210.9617490.9696430.9682270.9673460.97869300:03
340.1640400.1098690.9562840.9658830.9646550.9637390.97541400:03
350.1648270.1145080.9453550.9499020.9518130.9505260.97620600:03
360.1586140.1054670.9508200.9628040.9583740.9575900.97698700:03
370.1504520.1319240.9453550.9525870.9508560.9494440.97628100:03
380.1488820.1396440.9344260.9417810.9433160.9415680.96824900:03
390.1453530.1312800.9617490.9628470.9656340.9643480.97869800:03
400.1410540.1275070.9508200.9560570.9557820.9545360.97706300:03
410.1283400.1381920.9344260.9426520.9419610.9404470.97460700:03
420.1292190.1214620.9344260.9424300.9419610.9403650.97464500:03
430.1308900.1139400.9562840.9659800.9633010.9624000.97787500:03
440.1295620.1552430.9508200.9536630.9567400.9552960.97709400:03
450.1293250.1347720.9398910.9460230.9468870.9453970.97546100:04
460.1237180.1162030.9508200.9596610.9593320.9584160.97701900:03
470.1229690.1384540.9508200.9533730.9567400.9553570.97706000:03
480.1205360.1149380.9508200.9533730.9567400.9553570.97706000:03
490.1222910.1278290.9398910.9460870.9468870.9454720.97542400:03
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.994535505771637.\n", "Better model found at epoch 3 with accuracy value: 1.0.\n" ] } ], "source": [ "learner.fit_one_cycle(50, max_lr=slice(1e-05, 1e-04), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold4-stage2\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold4-stage2\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Fold-5" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "ImageDataBunch;\n", "\n", "Train: LabelList (734 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Valid: LabelList (183 items)\n", "x: ImageList\n", "Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64),Image (3, 64, 64)\n", "y: CategoryList\n", "abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic,abnormal_moderate-dysplastic\n", "Path: ../../../Dataset/Herlev Dataset;\n", "\n", "Test: None" ] }, "execution_count": 32, "metadata": {}, "output_type": "execute_result" } ], "source": [ "fold_idxs = idxs[4]\n", "fold_data = (ImageList.from_folder(data_path)\n", " .split_by_idxs(fold_idxs[0], fold_idxs[1])\n", " .label_from_folder()\n", " .transform(tfms, size=64)\n", " .databunch(bs=64)\n", " .normalize(imagenet_stats))\n", "fold_data" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 70.00% [7/10 00:15<00:06]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.424367#na#00:02
10.452629#na#00:02
20.458376#na#00:02
30.456805#na#00:02
40.473501#na#00:02
50.582748#na#00:02
61.141868#na#00:02

\n", "\n", "

\n", " \n", " \n", " 18.18% [2/11 00:01<00:05 1.4290]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold4-stage2\")\n", "learner.data = fold_data\n", "learner.freeze()\n", "learner = to_fp16(learner)\n", "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.5387070.0761760.9781420.9792190.9839440.9827660.97755800:03
10.4705610.0857420.9781420.9800110.9836390.9827620.97502800:03
20.4881600.0745730.9836070.9836730.9876070.9867070.97839100:03
30.4836060.0757580.9836070.9836730.9876070.9867070.97839100:03
40.4629340.0793740.9781420.9792190.9839440.9828170.97105000:03
50.4540590.0773160.9781420.9792190.9839440.9828170.97105000:03
60.4492910.0752520.9836070.9836730.9876070.9867070.97839100:03
70.4548980.0771340.9836070.9836730.9876070.9867070.97839100:03
80.4547560.0785460.9836070.9836730.9876070.9867070.97839100:03
90.4565380.0828970.9836070.9836730.9876070.9867070.97839100:03
100.4582780.0802260.9781420.9792190.9839440.9828170.97105000:03
110.4578900.0769510.9836070.9836730.9876070.9867070.97839100:03
120.4556680.0776100.9836070.9836730.9876070.9867070.97839100:03
130.4556420.0743420.9836070.9836730.9876070.9867070.97839100:03
140.4492520.0758960.9836070.9836730.9876070.9867070.97839100:03
150.4519990.0799830.9836070.9836730.9876070.9867070.97839100:03
160.4571090.0781960.9836070.9836730.9876070.9867070.97839100:03
170.4647130.0791520.9836070.9836730.9876070.9867070.97839100:03
180.4724680.0766290.9836070.9836730.9876070.9867070.97839100:03
190.4754180.0746010.9836070.9836730.9876070.9867070.97839100:03
200.4699340.0777570.9836070.9836730.9876070.9867070.97839100:03
210.4560420.0775020.9836070.9836730.9876070.9867070.97839100:03
220.4621940.0756110.9726780.9747640.9802810.9788450.97022100:03
230.4741110.0782230.9836070.9836730.9876070.9867070.97839100:03
240.4753320.1152420.9781420.9741500.9836390.9814670.96521600:03
250.4727930.0773900.9836070.9836730.9876070.9867070.97839100:03
260.4740300.0734260.9836070.9836730.9876070.9867070.97839100:03
270.4706560.0740830.9836070.9836730.9876070.9867070.97839100:03
280.4712480.0744190.9836070.9836730.9876070.9867070.97839100:03
290.4666590.0740510.9836070.9836730.9876070.9867070.97839100:03
300.4557800.0764240.9836070.9836730.9876070.9867070.97839100:03
310.4594850.0748580.9781420.9792190.9839440.9828170.97105000:03
320.4666330.0772950.9836070.9836730.9876070.9867070.97839100:03
330.4606410.0758100.9836070.9836730.9876070.9867070.97839100:03
340.4669150.0760340.9836070.9836730.9876070.9867070.97839100:03
350.4695320.0741960.9781420.9792190.9839440.9828170.97105000:03
360.4698220.0767260.9836070.9836730.9876070.9867070.97839100:03
370.4626140.0774600.9836070.9836730.9876070.9867070.97839100:03
380.4609750.0777710.9836070.9836730.9876070.9867070.97839100:03
390.4542350.0870800.9781420.9800110.9836390.9827620.97502800:03
400.4479080.0766210.9836070.9836730.9876070.9867070.97839100:03
410.4555510.0773240.9836070.9836730.9876070.9867070.97839100:03
420.4507010.0746440.9836070.9836730.9876070.9867070.97839100:03
430.4453980.0764640.9836070.9836730.9876070.9867070.97839100:03
440.4475960.0744190.9836070.9836730.9876070.9867070.97839100:03
450.4427910.0741700.9836070.9836730.9876070.9867070.97839100:03
460.4428920.0735420.9836070.9836730.9876070.9867070.97839100:03
470.4493970.0760220.9781420.9792190.9839440.9828170.97105000:03
480.4503500.0760720.9836070.9836730.9876070.9867070.97839100:03
490.4503380.0772590.9781420.9792190.9839440.9828170.97105000:03
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.9781420826911926.\n", "Better model found at epoch 2 with accuracy value: 0.9836065769195557.\n" ] } ], "source": [ "learner.fit_one_cycle(50, max_lr=slice(1e-06), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold5-stage1\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold5-stage1\")" ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 70.00% [7/10 00:17<00:07]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.493384#na#00:02
10.478736#na#00:02
20.473908#na#00:02
30.459177#na#00:02
40.673044#na#00:02
51.013287#na#00:02
61.689738#na#00:02

\n", "\n", "

\n", " \n", " \n", " 0.00% [0/11 00:00<00:00]\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold5-stage1\")\n", "learner = to_fp16(learner)\n", "learner.unfreeze()\n", "learner.lr_find()\n", "learner.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracyprecisionrecallf_betakappa_scoretime
00.4288330.0764740.9836070.9836730.9876070.9867070.97839100:03
10.4283440.0795770.9781420.9792190.9839440.9828170.97105000:03
20.4525850.0760680.9836070.9836730.9876070.9867070.97839100:03
30.4494930.0800850.9836070.9836730.9876070.9867070.97839100:03
40.4619200.0814190.9836070.9836730.9876070.9867070.97839100:03
50.4620760.0807570.9836070.9836730.9876070.9867070.97839100:03
60.4624440.0797570.9781420.9792190.9839440.9827660.97755800:03
70.4579440.0811950.9726780.9747640.9802810.9788450.97022100:03
80.4411750.0818440.9836070.9836730.9876070.9867070.97839100:03
90.4396180.0832020.9836070.9836730.9876070.9867070.97839100:03
100.4360390.0819440.9836070.9836730.9876070.9867070.97839100:03
110.4200870.0850900.9781420.9792190.9839440.9827660.97755800:03
120.4234600.0846910.9836070.9836730.9876070.9867070.97839100:03
130.4162600.0925620.9781420.9797050.9839440.9829550.97510500:03
140.4103220.0954810.9781420.9797050.9839440.9829550.97510500:03
150.4176790.0983370.9726780.9752950.9791820.9781930.96993300:03
160.4062930.0965830.9672130.9656280.9696580.9685290.96240900:03
170.3994880.0924740.9726780.9703220.9733210.9723890.96978100:03
180.3920530.0974150.9617490.9615590.9659950.9647580.95913100:03
190.3881150.0978970.9781420.9792190.9839440.9828170.97105000:03
200.3912400.0970100.9781420.9798460.9828450.9821230.97077000:03
210.3752500.0968500.9726780.9757870.9790180.9782400.97431600:03
220.3665840.0951830.9726780.9757820.9791820.9783830.96748000:03
230.3612400.1009500.9672130.9662580.9696580.9686480.96647600:03
240.3588630.1029120.9672130.9662580.9696580.9686480.96647600:03
250.3634220.1006900.9726780.9757820.9791820.9783830.96748000:03
260.3510100.0985750.9672130.9710820.9755190.9744920.96016700:03
270.3462910.1007510.9726780.9752950.9791820.9781930.96993300:03
280.3437340.1040960.9672130.9710200.9742560.9734900.96914700:03
290.3345050.0967540.9726780.9752950.9791820.9781930.96993300:03
300.3325060.0946610.9726780.9752950.9791820.9781930.96993300:03
310.3357700.1010050.9726780.9757820.9791820.9783830.96748000:03
320.3245910.1055020.9672130.9710200.9742560.9734900.96914700:03
330.3262710.1039060.9672130.9710200.9742560.9734900.96914700:03
340.3206360.1032680.9672130.9710200.9742560.9734900.96914700:03
350.3106250.1034920.9617490.9682060.9705930.9699140.96342600:03
360.3140580.1044310.9672130.9712250.9755190.9744220.96664400:03
370.3135850.1033610.9672130.9712250.9755190.9744220.96664400:03
380.3127020.1007390.9726780.9752950.9791820.9781930.96993300:03
390.3058820.1059080.9726780.9752950.9791820.9781930.96993300:03
400.3116380.0986840.9726780.9752950.9791820.9781930.96993300:03
410.3086790.1008460.9726780.9752950.9791820.9781930.96993300:03
420.3010790.1000470.9726780.9752950.9791820.9781930.96993300:03
430.2944490.1003820.9726780.9752950.9791820.9781930.96993300:03
440.2965120.0978830.9726780.9752950.9791820.9781930.96993300:03
450.2998810.1027100.9726780.9752950.9791820.9781930.96993300:03
460.2986970.1021950.9672130.9712250.9755190.9744220.96664400:03
470.2997750.1030540.9672130.9712250.9755190.9744220.96664400:03
480.2995960.1014320.9726780.9752950.9791820.9781930.96993300:03
490.3025230.0968090.9726780.9752950.9791820.9781930.96993300:03
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Better model found at epoch 0 with accuracy value: 0.9836065769195557.\n" ] } ], "source": [ "learner.fit_one_cycle(50, max_lr=slice(5e-05), callbacks=model_callback(learner, \"best-rn101-herlev-multiclass-fold5-stage2\"))\n", "learner.save(\"last-rn101-herlev-multiclass-fold5-stage2\")" ] }, { "cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [], "source": [ "learner.load(\"best-rn101-herlev-multiclass-fold5-stage2\")\n", "learner.export(\"best-rn101-herlev-multiclass-5fold.pkl\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.9" } }, "nbformat": 4, "nbformat_minor": 4 }