Noninteractive run of notebooks and flake8 checks (#107)

* start of automatic and update occ. sens. * current progress * current progress * current progress * current progress * vae * models ensemble * layer_wise_learning_rate * gan * 3d_classification * add 3d_segmentation as well as flake * transform speed * autopep8 * update runner * flake8 support added * max_num_epochs->max_epochs * check that max_epochs exists unless not expected * uncomment executing notebook * check pip install * ignore temp files * no noqa for indented import monai * dont check for indented import monai * flake8 changes * magic pip installs * autofixes * use ! instead of % for pip install * so far * current progress. add black, isort, autoflake * current progress * update class lung lesion notebook * class lung lesion * current progress * current progress * finished * finished * remove pip install of pinned pytorch version * = list() to = [] * [DLMED] fix dyunet notebook issue Signed-off-by: Nic Ma <nma@nvidia.com> * correct faulty import * pep8 for dynunet * all working * add missing quotations * remove personal file path * remove NiftiDataset * dont remove EOL whitespace from comments * 2d_classification * re-add spaces * spaces * last spaces * final changes * notification at end * notification on exit * add github action * data subfolder * make folder if necessary * add flake8 to requirements * change test name Co-authored-by: Nic Ma <nma@nvidia.com>
Project-MONAI · Jan 27, 2021 · 59b57a4 · 59b57a4
1 parent 49e78ff
commit 59b57a4
Show file tree

Hide file tree

Showing 36 changed files with 4,673 additions and 2,373 deletions.
diff --git a/.github/workflows/pep8.yml b/.github/workflows/pep8.yml
@@ -0,0 +1,39 @@
+name: build
+
+on:
+  # quick tests for every pull request
+  push:
+    branches:
+      - master
+  pull_request:
+
+jobs:
+  # caching of these jobs:
+  #   - docker-20-03-py3-pip- (shared)
+  #   - ubuntu py37 pip-
+  #   - os-latest-pip- (shared)
+  pep8:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v2
+    - name: Set up Python 3.8
+      uses: actions/setup-python@v2
+      with:
+        python-version: 3.8
+    - name: cache weekly timestamp
+      id: pip-cache
+      run: |
+        echo "::set-output name=datew::$(date '+%Y-%V')"
+    - name: cache for pip
+      uses: actions/cache@v2
+      id: cache
+      with:
+        path: ~/.cache/pip
+        key: ${{ runner.os }}-pip-${{ steps.pip-cache.outputs.datew }}
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip wheel
+        python -m pip install -r requirements.txt
+    - name: PEP8 check
+      run: |
+        $(pwd)/runner.sh --no-run
diff --git a/.gitignore b/.gitignore
@@ -110,6 +110,7 @@ temp/
 .idea/
 
 *~
+._*
 
 # Remove .pyre temporary config files
 .pyre
@@ -133,3 +134,7 @@ tests/testing_data/*Hippocampus*
 
 # saved networks
 *.pth
+
+# Ignore torch saves
+*/torch/runs
+logs
diff --git a/2d_classification/mednist_tutorial.ipynb b/2d_classification/mednist_tutorial.ipynb
@@ -31,21 +31,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "try:\n",
-    "    import monai\n",
-    "except ImportError:\n",
-    "    %pip install -q \"monai[pillow, tqdm]\""
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {
-    "scrolled": true,
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
+    "!python -c \"import monai\" || pip install -q \"monai[pillow, tqdm]\"\n",
+    "!python -c \"import matplotlib\" || pip install -q matplotlib\n",
     "%matplotlib inline"
    ]
   },
@@ -107,12 +94,12 @@
     "import os\n",
     "import shutil\n",
     "import tempfile\n",
-    "\n",
     "import matplotlib.pyplot as plt\n",
-    "import numpy as np\n",
     "import PIL\n",
     "import torch\n",
     "import numpy as np\n",
+    "from sklearn.metrics import classification_report\n",
+    "\n",
     "from monai.apps import download_and_extract\n",
     "from monai.config import print_config\n",
     "from monai.metrics import compute_roc_auc\n",
@@ -154,7 +141,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "/home/rbrown/data/MONAI\n"
+      "/workspace/data/medical\n"
      ]
     }
    ],
@@ -243,7 +230,8 @@
     }
    ],
    "source": [
-    "class_names = sorted(x for x in os.listdir(data_dir) if os.path.isdir(os.path.join(data_dir, x)))\n",
+    "class_names = sorted(x for x in os.listdir(data_dir)\n",
+    "                     if os.path.isdir(os.path.join(data_dir, x)))\n",
     "num_class = len(class_names)\n",
     "image_files = [\n",
     "    [\n",
@@ -331,12 +319,12 @@
    "source": [
     "val_frac = 0.1\n",
     "test_frac = 0.1\n",
-    "train_x = list()\n",
-    "train_y = list()\n",
-    "val_x = list()\n",
-    "val_y = list()\n",
-    "test_x = list()\n",
-    "test_y = list()\n",
+    "train_x = []\n",
+    "train_y = []\n",
+    "val_x = []\n",
+    "val_y = []\n",
+    "test_x = []\n",
+    "test_y = []\n",
     "\n",
     "for i in range(num_total):\n",
     "    rann = np.random.random()\n",
@@ -350,7 +338,9 @@
     "        train_x.append(image_files_list[i])\n",
     "        train_y.append(image_class[i])\n",
     "\n",
-    "print(f\"Training count: {len(train_x)}, Validation count: {len(val_x)}, Test count: {len(test_x)}\")"
+    "print(\n",
+    "    f\"Training count: {len(train_x)}, Validation count: \"\n",
+    "    f\"{len(val_x)}, Test count: {len(test_x)}\")"
    ]
   },
   {
@@ -378,7 +368,8 @@
     "    ]\n",
     ")\n",
     "\n",
-    "val_transforms = Compose([LoadImage(image_only=True), AddChannel(), ScaleIntensity(), ToTensor()])"
+    "val_transforms = Compose(\n",
+    "    [LoadImage(image_only=True), AddChannel(), ScaleIntensity(), ToTensor()])"
    ]
   },
   {
@@ -401,13 +392,16 @@
     "\n",
     "\n",
     "train_ds = MedNISTDataset(train_x, train_y, train_transforms)\n",
-    "train_loader = torch.utils.data.DataLoader(train_ds, batch_size=300, shuffle=True, num_workers=10)\n",
+    "train_loader = torch.utils.data.DataLoader(\n",
+    "    train_ds, batch_size=300, shuffle=True, num_workers=10)\n",
     "\n",
     "val_ds = MedNISTDataset(val_x, val_y, val_transforms)\n",
-    "val_loader = torch.utils.data.DataLoader(val_ds, batch_size=300, num_workers=10)\n",
+    "val_loader = torch.utils.data.DataLoader(\n",
+    "    val_ds, batch_size=300, num_workers=10)\n",
     "\n",
     "test_ds = MedNISTDataset(test_x, test_y, val_transforms)\n",
-    "test_loader = torch.utils.data.DataLoader(test_ds, batch_size=300, num_workers=10)"
+    "test_loader = torch.utils.data.DataLoader(\n",
+    "    test_ds, batch_size=300, num_workers=10)"
    ]
   },
   {
@@ -419,7 +413,7 @@
     "1. Set learning rate for how much the model is updated per batch.\n",
     "1. Set total epoch number, as we have shuffle and random transforms, so the training data of every epoch is different.  \n",
     "And as this is just a get start tutorial, let's just train 4 epochs.  \n",
-    "If train 10 epochs, the model can achieve 100% accuracy on test dataset.\n",
+    "If train 10 epochs, the model can achieve 100% accuracy on test dataset. \n",
     "1. Use DenseNet from MONAI and move to GPU devide, this DenseNet can support both 2D and 3D classification tasks.\n",
     "1. Use Adam optimizer."
    ]
@@ -431,10 +425,11 @@
    "outputs": [],
    "source": [
     "device = torch.device(\"cuda:0\")\n",
-    "model = densenet121(spatial_dims=2, in_channels=1, out_channels=num_class).to(device)\n",
+    "model = densenet121(spatial_dims=2, in_channels=1,\n",
+    "                    out_channels=num_class).to(device)\n",
     "loss_function = torch.nn.CrossEntropyLoss()\n",
     "optimizer = torch.optim.Adam(model.parameters(), 1e-5)\n",
-    "max_num_epochs = 4\n",
+    "max_epochs = 4\n",
     "val_interval = 1"
    ]
   },
@@ -1130,12 +1125,12 @@
    "source": [
     "best_metric = -1\n",
     "best_metric_epoch = -1\n",
-    "epoch_loss_values = list()\n",
-    "metric_values = list()\n",
+    "epoch_loss_values = []\n",
+    "metric_values = []\n",
     "\n",
-    "for epoch in range(max_num_epochs):\n",
+    "for epoch in range(max_epochs):\n",
     "    print(\"-\" * 10)\n",
-    "    print(f\"epoch {epoch + 1}/{max_num_epochs}\")\n",
+    "    print(f\"epoch {epoch + 1}/{max_epochs}\")\n",
     "    model.train()\n",
     "    epoch_loss = 0\n",
     "    step = 0\n",
@@ -1148,7 +1143,9 @@
     "        loss.backward()\n",
     "        optimizer.step()\n",
     "        epoch_loss += loss.item()\n",
-    "        print(f\"{step}/{len(train_ds) // train_loader.batch_size}, train_loss: {loss.item():.4f}\")\n",
+    "        print(\n",
+    "            f\"{step}/{len(train_ds) // train_loader.batch_size}, \"\n",
+    "            f\"train_loss: {loss.item():.4f}\")\n",
     "        epoch_len = len(train_ds) // train_loader.batch_size\n",
     "    epoch_loss /= step\n",
     "    epoch_loss_values.append(epoch_loss)\n",
@@ -1166,22 +1163,27 @@
     "                )\n",
     "                y_pred = torch.cat([y_pred, model(val_images)], dim=0)\n",
     "                y = torch.cat([y, val_labels], dim=0)\n",
-    "            auc_metric = compute_roc_auc(y_pred, y, to_onehot_y=True, softmax=True)\n",
+    "            auc_metric = compute_roc_auc(\n",
+    "                y_pred, y, to_onehot_y=True, softmax=True)\n",
     "            metric_values.append(auc_metric)\n",
     "            acc_value = torch.eq(y_pred.argmax(dim=1), y)\n",
     "            acc_metric = acc_value.sum().item() / len(acc_value)\n",
     "            if auc_metric > best_metric:\n",
     "                best_metric = auc_metric\n",
     "                best_metric_epoch = epoch + 1\n",
-    "                torch.save(model.state_dict(), os.path.join(root_dir, \"best_metric_model.pth\"))\n",
+    "                torch.save(model.state_dict(), os.path.join(\n",
+    "                    root_dir, \"best_metric_model.pth\"))\n",
     "                print(\"saved new best metric model\")\n",
     "            print(\n",
     "                f\"current epoch: {epoch + 1} current AUC: {auc_metric:.4f}\"\n",
-    "                f\" current accuracy: {acc_metric:.4f} best AUC: {best_metric:.4f}\"\n",
+    "                f\" current accuracy: {acc_metric:.4f}\"\n",
+    "                f\" best AUC: {best_metric:.4f}\"\n",
     "                f\" at epoch: {best_metric_epoch}\"\n",
     "            )\n",
     "\n",
-    "print(f\"train completed, best_metric: {best_metric:.4f} at epoch: {best_metric_epoch}\")"
+    "print(\n",
+    "    f\"train completed, best_metric: {best_metric:.4f} \"\n",
+    "    f\"at epoch: {best_metric_epoch}\")"
    ]
   },
   {
@@ -1243,10 +1245,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model.load_state_dict(torch.load(os.path.join(root_dir, \"best_metric_model.pth\")))\n",
+    "model.load_state_dict(torch.load(\n",
+    "    os.path.join(root_dir, \"best_metric_model.pth\")))\n",
     "model.eval()\n",
-    "y_true = list()\n",
-    "y_pred = list()\n",
+    "y_true = []\n",
+    "y_pred = []\n",
     "with torch.no_grad():\n",
     "    for test_data in test_loader:\n",
     "        test_images, test_labels = (\n",
@@ -1288,13 +1291,8 @@
     }
    ],
    "source": [
-    "try:\n",
-    "    import sklean\n",
-    "except ImportError:\n",
-    "    %pip install -qU sklearn\n",
-    "from sklearn.metrics import classification_report\n",
-    "\n",
-    "print(classification_report(y_true, y_pred, target_names=class_names, digits=4))"
+    "print(classification_report(\n",
+    "    y_true, y_pred, target_names=class_names, digits=4))"
    ]
   },
   {