updated AMP tutorial for 0.9.1rc2

Author: Yuchen Xu

acceleration/automatic_mixed_precision.ipynb

@@ -28,17 +28,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Note: you may need to restart the kernel to use updated packages.\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "!python -c \"import monai\" || pip install -q \"monai-weekly[nibabel, tqdm]\"\n",
     "!python -c \"import matplotlib\" || pip install -q matplotlib\n",
@@ -54,40 +46,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "MONAI version: 0.6.0rc1+15.gf3d436a0\n",
-      "Numpy version: 1.20.3\n",
-      "Pytorch version: 1.9.0a0+c3d40fd\n",
-      "MONAI flags: HAS_EXT = True, USE_COMPILED = False\n",
-      "MONAI rev id: f3d436a09deefcf905ece2faeec37f55ab030003\n",
-      "\n",
-      "Optional dependencies:\n",
-      "Pytorch Ignite version: 0.4.5\n",
-      "Nibabel version: 3.2.1\n",
-      "scikit-image version: 0.15.0\n",
-      "Pillow version: 8.2.0\n",
-      "Tensorboard version: 2.5.0\n",
-      "gdown version: 3.13.0\n",
-      "TorchVision version: 0.10.0a0\n",
-      "ITK version: 5.1.2\n",
-      "tqdm version: 4.53.0\n",
-      "lmdb version: 1.2.1\n",
-      "psutil version: 5.8.0\n",
-      "pandas version: 1.1.4\n",
-      "einops version: 0.3.0\n",
-      "\n",
-      "For details about installing the optional dependencies, please visit:\n",
-      "    https://docs.monai.io/en/latest/installation.html#installing-the-recommended-dependencies\n",
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# Copyright 2020 MONAI Consortium\n",
     "# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
@@ -154,17 +115,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "root dir is: /workspace/data/medical\n"
-     ]
-    }
-   ],
+   "execution_count": null,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [],
    "source": [
     "directory = os.environ.get(\"MONAI_DATA_DIRECTORY\")\n",
     "root_dir = tempfile.mkdtemp() if directory is None else directory\n",
@@ -182,7 +137,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -204,7 +159,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -230,7 +185,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -275,7 +230,6 @@
     "                fg_indices_key=\"label_fg\",\n",
     "                bg_indices_key=\"label_bg\",\n",
     "            ),\n",
-    "            EnsureTyped(keys=[\"image\", \"label\"]),\n",
     "        ]\n",
     "    )\n",
     "    val_transforms = Compose(\n",
@@ -297,7 +251,6 @@
     "                clip=True,\n",
     "            ),\n",
     "            CropForegroundd(keys=[\"image\", \"label\"], source_key=\"image\"),\n",
-    "            EnsureTyped(keys=[\"image\", \"label\"]),\n",
     "        ]\n",
     "    )\n",
     "    return train_transforms, val_transforms"
@@ -313,7 +266,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "metadata": {
     "scrolled": true
    },
@@ -352,8 +305,8 @@
     "    optimizer = torch.optim.Adam(model.parameters(), 1e-4)\n",
     "    scaler = torch.cuda.amp.GradScaler() if amp else None\n",
     "\n",
-    "    post_pred = Compose([EnsureType(), AsDiscrete(argmax=True, to_onehot=2)])\n",
-    "    post_label = Compose([EnsureType(), AsDiscrete(to_onehot=2)])\n",
+    "    post_pred = Compose([AsDiscrete(argmax=True, to_onehot=2)])\n",
+    "    post_label = Compose([AsDiscrete(to_onehot=2)])\n",
     "\n",
     "    dice_metric = DiceMetric(include_background=False, reduction=\"mean\", get_not_nans=False)\n",
     "\n",
@@ -422,8 +375,8 @@
     "                        val_outputs = sliding_window_inference(\n",
     "                            val_inputs, roi_size, sw_batch_size, model\n",
     "                        )\n",
-    "                    val_outputs = [post_pred(i) for i in decollate_batch(val_outputs)]\n",
-    "                    val_labels = [post_label(i) for i in decollate_batch(val_labels)]\n",
+    "                    val_outputs = [post_pred(i) for i in decollate_batch(val_outputs.as_tensor())]\n",
+    "                    val_labels = [post_label(i) for i in decollate_batch(val_labels.as_tensor())]\n",
     "                    dice_metric(y_pred=val_outputs, y=val_labels)\n",
     "\n",
     "                metric = dice_metric.aggregate().item()\n",
@@ -474,7 +427,9 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "metadata": {},
+   "metadata": {
+    "scrolled": true
+   },
    "outputs": [],
    "source": [
     "set_determinism(seed=0)\n",
@@ -856,7 +811,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -870,7 +825,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.10"
+   "version": "3.8.13"
   }
  },
  "nbformat": 4,