[GraphBolt] move to_dgl() for tutorial notebooks (#6727)

notebooks/stochastic_training/link_prediction.ipynb

@@ -4,9 +4,7 @@
   "metadata": {
     "colab": {
       "private_outputs": true,
-      "provenance": [],
-      "authorship_tag": "ABX9TyOjqI7Q6kAUIF+Fhf3q8KUM",
-      "include_colab_link": true
+      "provenance": []
     },
     "kernelspec": {
       "name": "python3",
@@ -143,7 +141,6 @@
         "datapipe = datapipe.sample_uniform_negative(graph, 5)\n",
         "datapipe = datapipe.sample_neighbor(graph, [5, 5, 5])\n",
         "datapipe = datapipe.fetch_feature(feature, node_feature_keys=[\"feat\"])\n",
-        "datapipe = datapipe.to_dgl()\n",
         "datapipe = datapipe.copy_to(device)\n",
         "train_dataloader = gb.DataLoader(datapipe, num_workers=0)"
       ],
@@ -167,7 +164,7 @@
       "cell_type": "code",
       "source": [
         "data = next(iter(train_dataloader))\n",
-        "print(f\"DGLMiniBatch: {data}\")"
+        "print(f\"MiniBatch: {data}\")"
       ],
       "metadata": {
         "id": "euEdzmerYmZi"
@@ -175,6 +172,27 @@
       "execution_count": null,
       "outputs": []
     },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "In order to train with DGL, you need to convert `MiniBatch` to `DGLMiniBatch` like below:"
+      ],
+      "metadata": {
+        "id": "IpAgrEp_cdEP"
+      }
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "data = data.to_dgl()\n",
+        "print(f\"DGLMiniBatch: {data}\")"
+      ],
+      "metadata": {
+        "id": "KQgxFUyCcjVT"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
     {
       "cell_type": "markdown",
       "source": [
@@ -295,7 +313,10 @@
         "    model.train()\n",
         "    total_loss = 0\n",
         "    for step, data in tqdm.tqdm(enumerate(train_dataloader)):\n",
-        "        # Unpack MiniBatch.\n",
+        "        # Convert to DGL format.\n",
+        "        data = data.to_dgl()\n",
+        "\n",
+        "        # Unpack DGLMiniBatch.\n",
         "        compacted_pairs, labels = to_binary_link_dgl_computing_pack(data)\n",
         "        node_feature = data.node_features[\"feat\"]\n",
         "        # Convert sampled subgraphs to DGL blocks.\n",
@@ -342,13 +363,15 @@
         "# to -1.\n",
         "datapipe = datapipe.sample_neighbor(graph, [-1, -1])\n",
         "datapipe = datapipe.fetch_feature(feature, node_feature_keys=[\"feat\"])\n",
-        "datapipe = datapipe.to_dgl()\n",
         "datapipe = datapipe.copy_to(device)\n",
         "eval_dataloader = gb.DataLoader(datapipe, num_workers=0)\n",
         "\n",
         "logits = []\n",
         "labels = []\n",
         "for step, data in enumerate(eval_dataloader):\n",
+        "    # Convert to DGL format.\n",
+        "    data = data.to_dgl()\n",
+        "\n",
         "    # Unpack MiniBatch.\n",
         "    compacted_pairs, label = to_binary_link_dgl_computing_pack(data)\n",
         "\n",
@@ -393,4 +416,4 @@
       }
     }
   ]
-}
+}
\ No newline at end of file

notebooks/stochastic_training/node_classification.ipynb

@@ -152,7 +152,6 @@
         "datapipe = gb.ItemSampler(train_set, batch_size=1024, shuffle=True)\n",
         "datapipe = datapipe.sample_neighbor(graph, [4, 4])\n",
         "datapipe = datapipe.fetch_feature(feature, node_feature_keys=[\"feat\"])\n",
-        "datapipe = datapipe.to_dgl()\n",
         "datapipe = datapipe.copy_to(device)\n",
         "train_dataloader = gb.DataLoader(datapipe, num_workers=0)"
       ],
@@ -165,7 +164,7 @@
     {
       "cell_type": "markdown",
       "source": [
-        "You can iterate over the data loader and a `DGLMiniBatch` object is yielded.\n",
+        "You can iterate over the data loader and a `MiniBatch` object is yielded.\n",
         "\n"
       ],
       "metadata": {
@@ -184,6 +183,27 @@
       "execution_count": null,
       "outputs": []
     },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "In order to train with DGL, you need to convert `MiniBatch` to `DGLMiniBatch` like below:"
+      ],
+      "metadata": {
+        "id": "FwDJf1AJbNtt"
+      }
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "data = data.to_dgl()\n",
+        "print(data)"
+      ],
+      "metadata": {
+        "id": "3Tzfp6A8bdWv"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
     {
       "cell_type": "markdown",
       "source": [
@@ -285,7 +305,6 @@
         "datapipe = gb.ItemSampler(valid_set, batch_size=1024, shuffle=False)\n",
         "datapipe = datapipe.sample_neighbor(graph, [4, 4])\n",
         "datapipe = datapipe.fetch_feature(feature, node_feature_keys=[\"feat\"])\n",
-        "datapipe = datapipe.to_dgl()\n",
         "datapipe = datapipe.copy_to(device)\n",
         "valid_dataloader = gb.DataLoader(datapipe, num_workers=0)\n",
         "\n",
@@ -317,6 +336,7 @@
         "\n",
         "    with tqdm.tqdm(train_dataloader) as tq:\n",
         "        for step, data in enumerate(tq):\n",
+        "            data = data.to_dgl()\n",
         "            x = data.node_features[\"feat\"]\n",
         "            labels = data.labels\n",
         "\n",
@@ -343,6 +363,7 @@
         "    labels = []\n",
         "    with tqdm.tqdm(valid_dataloader) as tq, torch.no_grad():\n",
         "        for data in tq:\n",
+        "            data = data.to_dgl()\n",
         "            x = data.node_features[\"feat\"]\n",
         "            labels.append(data.labels.cpu().numpy())\n",
         "            predictions.append(model(data.blocks, x).argmax(1).cpu().numpy())\n",