XCS224W_Colab4.ipynb

{"cells":[{"cell_type":"markdown","metadata":{"id":"XuXWJLEm2UWS"},"source":["# **CS224W - Colab 4**"]},{"cell_type":"markdown","metadata":{"id":"8gzsP50bF6Gb"},"source":["In this Colab, we shift our focus from homogenous graphs to heterogeneous graphs. Heterogeneous graphs extend the traditional homogenous graphs that we have been working with by incorporating **different node and edge types**. This additional information allows us to extend the graph neural nework models that we have worked with before. Namely, we can apply heterogenous message passing, where different message types now exist between different node and edge type relationships. \n","\n","In this notebook, you will first learn how to transform NetworkX graphs into DeepSNAP representations. Then, you will dive deeper into how DeepSNAP stores and represents heterogeneous graphs as PyTorch Tensors.\n","\n","With this knowledge, you will build your own heterogenous graph neural netowrk models using PyTorch Geometric and DeepSNAP. You will then apply your models for a node property prediction task; specifically, you will evaluate these models on the heterogeneous ACM node prediction dataset.\n","\n","Lastly, we give a sneak preview into the work you will do on Colab5. To close out this assignment we introduce the concept of Neighborhood Sampling and mini-batch training for scaling GNNs to large graphs.\n","\n","**Note**: Make sure to **sequentially run all the cells in each section**, so that the intermediate variables / packages will carry over to the next cell\n","\n","Have fun and good luck on Colab 4 :)"]},{"cell_type":"markdown","metadata":{"id":"MSaetj53YnT6"},"source":["# Device\n","You might need to use GPU for this Colab.\n","\n","Please click `Runtime` and then `Change runtime type`. Then set the `hardware accelerator` to **GPU**."]},{"cell_type":"markdown","metadata":{"id":"67gOQITlCNQi"},"source":["# Installation"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":14073,"status":"ok","timestamp":1657145036446,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"J_m9l6OYCQZP","outputId":"c59ce779-9b77-47fa-c04b-a3a2a7b332bc"},"outputs":[{"output_type":"stream","name":"stdout","text":["Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n","Looking in links: https://pytorch-geometric.com/whl/torch-1.11.0+cu113.html\n","Requirement already satisfied: torch-scatter in /usr/local/lib/python3.7/dist-packages (2.0.9)\n","Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n","Looking in links: https://pytorch-geometric.com/whl/torch-1.11.0+cu113.html\n","Requirement already satisfied: torch-sparse in /usr/local/lib/python3.7/dist-packages (0.6.14)\n","Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from torch-sparse) (1.4.1)\n","Requirement already satisfied: numpy>=1.13.3 in /usr/local/lib/python3.7/dist-packages (from scipy->torch-sparse) (1.21.6)\n","Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n","Requirement already satisfied: torch-geometric in /usr/local/lib/python3.7/dist-packages (2.0.4)\n","Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from torch-geometric) (1.21.6)\n","Requirement already satisfied: pyparsing in /usr/local/lib/python3.7/dist-packages (from torch-geometric) (3.0.9)\n","Requirement already satisfied: jinja2 in /usr/local/lib/python3.7/dist-packages (from torch-geometric) (2.11.3)\n","Requirement already satisfied: scikit-learn in /usr/local/lib/python3.7/dist-packages (from torch-geometric) (1.0.2)\n","Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from torch-geometric) (2.23.0)\n","Requirement already satisfied: tqdm in /usr/local/lib/python3.7/dist-packages (from torch-geometric) (4.64.0)\n","Requirement already satisfied: pandas in /usr/local/lib/python3.7/dist-packages (from torch-geometric) (1.3.5)\n","Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from torch-geometric) (1.4.1)\n","Requirement already satisfied: MarkupSafe>=0.23 in /usr/local/lib/python3.7/dist-packages (from jinja2->torch-geometric) (2.0.1)\n","Requirement already satisfied: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas->torch-geometric) (2022.1)\n","Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas->torch-geometric) (2.8.2)\n","Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas->torch-geometric) (1.15.0)\n","Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->torch-geometric) (2022.6.15)\n","Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->torch-geometric) (1.24.3)\n","Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->torch-geometric) (2.10)\n","Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->torch-geometric) (3.0.4)\n","Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->torch-geometric) (3.1.0)\n","Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->torch-geometric) (1.1.0)\n"]}],"source":["# Install torch geometric\n","import os\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  !pip install torch-scatter -f https://pytorch-geometric.com/whl/torch-1.11.0+cu113.html\n","  !pip install torch-sparse -f https://pytorch-geometric.com/whl/torch-1.11.0+cu113.html\n","  !pip install torch-geometric\n","  !pip install -q git+https://github.com/snap-stanford/deepsnap.git\n","  !pip install -U -q PyDrive"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":1146,"status":"ok","timestamp":1657145037581,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"Qpr0ThDgZmZV","outputId":"efdc8cf0-28df-49c7-9d68-71d76b936b69"},"outputs":[{"output_type":"stream","name":"stdout","text":["nvcc: NVIDIA (R) Cuda compiler driver\n","Copyright (c) 2005-2020 NVIDIA Corporation\n","Built on Mon_Oct_12_20:09:46_PDT_2020\n","Cuda compilation tools, release 11.1, V11.1.105\n","Build cuda_11.1.TC455_06.29190527_0\n","11.3\n"]}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  !nvcc --version\n","  !python -c \"import torch; print(torch.version.cuda)\""]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":7,"status":"ok","timestamp":1657145040262,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"PRfgbfTjCRD_","outputId":"58474b33-2248-455c-eb87-19c29805bae0"},"outputs":[{"output_type":"stream","name":"stdout","text":["1.11.0+cu113\n","2.0.4\n"]}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  import torch\n","  import torch_geometric\n","  print(torch.__version__)\n","  print(torch_geometric.__version__)"]},{"cell_type":"markdown","metadata":{"id":"ZoXlf4MtYrbz"},"source":["# 1) DeepSNAP Heterogeneous Graph\n","\n","First, you will explore how to transform a NetworkX graph into the format supported by DeepSNAP. \n","\n","DeepSNAP extends its traditional graph representation to include heterogeneous graphs by including the following graph property features:  \n","* `node_feature`: The feature of each node (`torch.tensor`)\n","* `edge_feature`: The feature of each edge (`torch.tensor`)\n","* `node_label`: The label of each node (`int`)\n","* `node_type`: The type of each node (`string`)\n","* `edge_type`: The type of each edge (`string`)\n","\n","where the key **new** features added are `node_type` and `edge_type`, which enables us to perform **heterogenous message passing**.\n","\n","For this first question you will work with the familiar [karate club graph](https://networkx.github.io/documentation/stable/auto_examples/graph/plot_karate_club.html) seen in Colab 1. To start, since each node in the graph belongs to one of two clubs (club \"Mr. Hi\" or club \"Officer\"), you will treat the club as the `node_type`. The code below demonstrates how to differentiate the nodes in the NetworkX graph.\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":535},"executionInfo":{"elapsed":893,"status":"ok","timestamp":1657145043888,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"8LQ_z5gcBVA1","outputId":"bffb54d5-73cb-400e-9364-028c95e82011"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 504x504 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["import networkx as nx\n","from networkx.algorithms.community import greedy_modularity_communities\n","import matplotlib.pyplot as plt\n","import copy\n","\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  from pylab import show\n","  G = nx.karate_club_graph()\n","  community_map = {}\n","  for node in G.nodes(data=True): # for each node, map its club type to 0 / 1 label to community_map dictionary\n","    if node[1][\"club\"] == \"Mr. Hi\": # if its attribute club = \"Mr.Hi\"    \n","      community_map[node[0]] = 0 # node[0] is node id, node[1] is a dictionary of node attributes\n","      #print(node)\n","    else:\n","      community_map[node[0]] = 1\n","  # print(community_map)\n","  node_color = []\n","  color_map = {0: 0, 1: 1}\n","  node_color = [color_map[community_map[node]] for node in G.nodes()]\n","  pos = nx.spring_layout(G)\n","  plt.figure(figsize=(7, 7))\n","  nx.draw(G, pos=pos, cmap=plt.get_cmap('coolwarm'), node_color=node_color)\n","  show()"]},{"cell_type":"markdown","metadata":{"id":"1JpFb9fTw1lg"},"source":["### Question 1.1: Assigning Node Type and Node Features (10 points)\n","\n","Using the `community_map` dictionary and graph `G` from above, add node attributes `node_type` and `node_label` to the graph G. Namely, for `node_type` assign nodes in the \"Mr. Hi\" club to a node type `n0` and nodes in club \"Officer\" a node type `n1`. Note: the node type should be a `string` property.\n","\n","Then for `node_label`, assign nodes in \"Mr. Hi\" club to a `node_label` `0` and nodes in club \"Officer\" a `node_label` of `1`.\n","\n","Lastly, assign every node the *tensor* feature vector $[1, 1, 1, 1, 1]$. \n","\n","**Hint**: Look at the NetworkX function `nx.classes.function.set_node_attributes`.\n","\n","**Note**: This question is not specifically graded but is important for later questions."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":224,"status":"ok","timestamp":1657145126556,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"zev_hMJHJXK1","outputId":"642b2f10-8924-4e67-98e1-15536cd13dfc"},"outputs":[{"output_type":"stream","name":"stdout","text":["Node 20 has properties: {'club': 'Officer', 'node_type': 'n1', 'node_label': 1, 'node_feature': tensor([1., 1., 1., 1., 1.])}\n"]}],"source":["import torch\n","\n","def assign_node_types(G, community_map):\n","  # TODO: Implement a function that takes in a NetworkX graph\n","  # G and community map assignment (mapping node id --> 0/1 label)\n","  # and adds 'node_type' as a node_attribute in G.\n","\n","  ############# Your code here ############\n","  ## (~2 line of code)\n","  ## Note\n","  ## 1. Look up NetworkX `nx.classes.function.set_node_attributes`\n","  ## 2. Look above for the two node type values!\n","  node_type_map = {0:'n0', 1:'n1'} # map node type from int to string\n","  node_types = {node:node_type_map[community_map[node]] for node in G.nodes()} # create a new dictionary {key=node, values=\"n0\"or \"n1\"}\n","  nx.classes.function.set_node_attributes(G, values=node_types, name='node_type')\n","  #########################################\n","\n","def assign_node_labels(G, community_map):\n","  # TODO: Implement a function that takes in a NetworkX graph\n","  # G and community map assignment (mapping node id --> 0/1 label)\n","  # and adds 'node_label' as a node_attribute in G.\n","\n","  ############# Your code here ############\n","  ## (~2 line of code)\n","  ## Note\n","  ## 1. Look up NetworkX `nx.classes.function.set_node_attributes`\n","  node_label_map = {0:0, 1:1}\n","  node_labels = {node:node_label_map[community_map[node]] for node in G.nodes()}\n","  nx.classes.function.set_node_attributes(G, values=node_labels, name='node_label')\n","  #########################################\n","\n","def assign_node_features(G):\n","  # TODO: Implement a function that takes in a NetworkX graph\n","  # G and adds 'node_feature' as a node_attribute in G. Each node\n","  # in the graph has the same feature vector - a torchtensor with \n","  # data [1., 1., 1., 1., 1.]\n","\n","  ############# Your code here ############\n","  ## (~2 line of code)\n","  ## Note\n","  ## 1. Look up NetworkX `nx.classes.function.set_node_attributes`\n","  node_features=torch.ones(5)\n","  nx.classes.function.set_node_attributes(G, values=node_features, name='node_feature')\n","  #########################################\n","\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  assign_node_types(G, community_map)\n","  assign_node_labels(G, community_map)\n","  assign_node_features(G)\n","\n","  # Explore node properties for the node with id: 20\n","  node_id = 20\n","  print (f\"Node {node_id} has properties:\", G.nodes(data=True)[node_id])"]},{"cell_type":"markdown","metadata":{"id":"mafN0P3EOhSb"},"source":["### Question 1.2: Assigning Edge Types (2.5 points)\n","\n","Next, we will assign three different `edge_types`: \n","* Edges within club \"Mr. Hi\": `e0`\n","* Edges within club \"Officer\": `e1`\n","* Edges between the two clubs: `e2`\n","\n","**Hint**: Use the `community_map` from before and `nx.classes.function.set_edge_attributes`"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":653,"status":"ok","timestamp":1657143099392,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"rsbYWEVwSV5n","outputId":"8e447816-a550-4361-a3e9-e0f48a0b83bb"},"outputs":[{"output_type":"stream","name":"stdout","text":["Edge (0, 31) has properties: {'edge_type': 'e2'}\n","Node 0 has properties: {'club': 'Mr. Hi', 'node_type': 'n0', 'node_label': 0, 'node_features': [1.0, 1.0, 1.0, 1.0, 1.0]}\n","Node 31 has properties: {'club': 'Officer', 'node_type': 'n1', 'node_label': 1, 'node_features': [1.0, 1.0, 1.0, 1.0, 1.0]}\n"]}],"source":["def assign_edge_types(G, community_map):\n","  # TODO: Implement a function that takes in a NetworkX graph\n","  # G and community map assignment (mapping node id --> 0/1 label)\n","  # and adds 'edge_type' as a edge_attribute in G.\n","\n","  ############# Your code here ############\n","  ## (~5 line of code)\n","  ## Note\n","  ## 1. Create an edge assignment dict following rules above\n","  ## 2. Look up NetworkX `nx.classes.function.set_edge_attributes`\n","  edge_type_map = {(0,0):'e0', (1,1):'e1', (0,1):'e2', (1,0):'e2'} # map community node type to edge type\n","  #print((community_map[edge[0]], community_map[edge[1]]))\n","  edge_types = {edge:edge_type_map[(community_map[edge[0]], community_map[edge[1]])] for edge in G.edges()} # create a new dictionary {key=node, values=\"n0\"or \"n1\"}\n","  #print(edge_types)\n","  nx.classes.function.set_edge_attributes(G, values=edge_types, name='edge_type')\n","  #########################################\n","\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  assign_edge_types(G, community_map)\n","\n","  # Explore edge properties for a sampled edge and check the corresponding\n","  # node types\n","  edge_idx = 15\n","  n1 = 0\n","  n2 = 31\n","  edge = list(G.edges(data=True))[edge_idx]\n","  print (f\"Edge ({edge[0]}, {edge[1]}) has properties:\", edge[2])\n","  print (f\"Node {n1} has properties:\", G.nodes(data=True)[n1])\n","  print (f\"Node {n2} has properties:\", G.nodes(data=True)[n2])"]},{"cell_type":"markdown","metadata":{"id":"VBsTdTPVTQ52"},"source":["## Heterogeneous Graph Visualization\n","\n","Now we can visualize the Heterogeneous Graph we have generated."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":607},"executionInfo":{"elapsed":35,"status":"ok","timestamp":1657143099393,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"n2sdufbODHtp","outputId":"cfa1bf33-37bf-4cc9-b30d-4eb8415a039b"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 576x576 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  edge_color = {}\n","  for edge in G.edges():\n","    n1, n2 = edge\n","    edge_color[edge] = community_map[n1] if community_map[n1] == community_map[n2] else 2\n","    if community_map[n1] == community_map[n2] and community_map[n1] == 0:\n","      edge_color[edge] = 'blue'\n","    elif community_map[n1] == community_map[n2] and community_map[n1] == 1:\n","      edge_color[edge] = 'red'\n","    else:\n","      edge_color[edge] = 'green'\n","\n","  G_orig = copy.deepcopy(G)\n","  nx.classes.function.set_edge_attributes(G, edge_color, name='color')\n","  colors = nx.get_edge_attributes(G,'color').values()\n","  labels = nx.get_node_attributes(G, 'node_type')\n","  plt.figure(figsize=(8, 8))\n","  nx.draw(G, pos=pos, cmap=plt.get_cmap('coolwarm'), node_color=node_color, edge_color=colors, labels=labels, font_color='white')\n","  show()"]},{"cell_type":"markdown","metadata":{"id":"QRANkF1jRxLV"},"source":["You should see that we differentiate edges within each clubs (2 types) and edges between the two clubs (1 type). Different types of nodes and edges are visualized in different colors. The NetworkX object `G` in following code can be transformed into `deepsnap.hetero_graph.HeteroGraph` directly."]},{"cell_type":"markdown","metadata":{"id":"4DW8L0hxbxw4"},"source":["## Transforming to DeepSNAP representation\n","\n","You will now work through transforming the **NetworkX object** `G` into a `deepsnap.hetero_graph.HeteroGraph`."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"-ZfHZ_eoVVGd"},"outputs":[],"source":["from deepsnap.hetero_graph import HeteroGraph\n","\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  hete = HeteroGraph(G_orig) "]},{"cell_type":"markdown","metadata":{"id":"izq4t_O9WxDH"},"source":["## Question 1.3: How many nodes are of each type (2.5 Points)\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":32,"status":"ok","timestamp":1657143099394,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"wRNSP6nnW78C","outputId":"f7de8cac-15ea-478f-8dc6-daec80246113"},"outputs":[{"output_type":"stream","name":"stdout","text":["Node type n0 has 17 nodes\n","Node type n1 has 17 nodes\n"]}],"source":["def get_nodes_per_type(hete):\n","  # TODO: Implement a function that takes a DeepSNAP dataset object\n","  # and return the number of nodes per `node_type`.\n","\n","  num_nodes_n0 = 0\n","  num_nodes_n1 = 0\n","\n","  ############# Your code here ############\n","  ## (~2 line of code)\n","  ## Note\n","  ## 1. Colab autocomplete functionality might be useful.\n","  num_nodes_n0 = hete.num_nodes(node_type=\"n0\")\n","  num_nodes_n1 = hete.num_nodes(node_type=\"n1\")\n","  #########################################\n","\n","  return num_nodes_n0, num_nodes_n1\n","\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  num_nodes_n0, num_nodes_n1 = get_nodes_per_type(hete)\n","  print(\"Node type n0 has {} nodes\".format(num_nodes_n0))\n","  print(\"Node type n1 has {} nodes\".format(num_nodes_n1))"]},{"cell_type":"markdown","metadata":{"id":"vEsHJp2ZYaE2"},"source":["## Question 1.4: Message Types - How many edges are of each message type (2.5 Points)\n","\n","When working with heterogenous graphs, we now have heterogenous message types (i.e. different message types for the different `node_type` and `edge_type` combinations). For example, an edge of type `e0` connecting two nodes in club \"Mr. HI\" would have a message type of (`n0`, `e0`, `n0`). In this problem we will analyze how many edges in our graph are of each message type.\n","\n","**Hint**: If you want to learn more about what the different message types are try the call `hete.message_types`"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":21,"status":"ok","timestamp":1657143099394,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"qobKuqbAYvJ7","outputId":"c9222223-1dff-4245-bb14-b40a1ff110a7"},"outputs":[{"output_type":"stream","name":"stdout","text":["Message type ('n0', 'e0', 'n0') has 35 edges\n","Message type ('n0', 'e2', 'n1') has 11 edges\n","Message type ('n1', 'e1', 'n1') has 32 edges\n"]}],"source":["def get_num_message_edges(hete):\n","  # TODO: Implement this function that takes a DeepSNAP dataset object\n","  # and return the number of edges for each message type. \n","  # You should return a list of tuples as \n","  # (message_type, num_edge)\n","\n","  message_type_edges = []\n","\n","  ############# Your code here ############\n","  ## (~2 line of code)\n","  ## Note\n","  ## 1. Colab autocomplete functionality might be useful.\n","  message_type_edges = hete.num_edges(message_type=hete.message_types).items()\n","  #########################################\n","\n","  return message_type_edges\n","\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  message_type_edges = get_num_message_edges(hete)\n","  for (message_type, num_edges) in message_type_edges:\n","    print(\"Message type {} has {} edges\".format(message_type, num_edges))"]},{"cell_type":"markdown","metadata":{"id":"bjMVik1JbJ76"},"source":["## Question 1.5: Dataset Splitting - How many nodes are in each dataset split? (2.5 Points)\n","\n","DeepSNAP has built in Dataset creation and splitting methods for heterogeneous graphs. Here you will create train, validation, and test datasets for a node prediction task and inspect the resulting subgraphs. Specifically, write a function that computes **the number of nodes with a known label in each dataset split.**\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":14,"status":"ok","timestamp":1657143099395,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"Ct10Oh4gcqgD","outputId":"f3c91b4b-b8ea-434b-a032-b66c8703e8ee"},"outputs":[{"output_type":"stream","name":"stdout","text":["train dataset has 12 nodes\n","val dataset has 10 nodes\n","test dataset has 12 nodes\n"]}],"source":["from deepsnap.dataset import GraphDataset\n","\n","def compute_dataset_split_counts(datasets):\n","  # TODO: Implement a function that takes a dict of datasets in the form\n","  # {'train': dataset_train, 'val': dataset_val, 'test': dataset_test}\n","  # and returns a dict mapping dataset names to the number of labeled\n","  # nodes used for supervision in that respective dataset.  \n","  \n","  data_set_splits = {}\n","\n","  ############# Your code here ############\n","  ## (~3 line of code)\n","  ## Note\n","  ## 1. The DeepSNAP `node_label_index` dictionary will be helpful. The keys are different labels, the values are index tensors of the nodes \n","  ## 2. Remember to count both node_types\n","  ## 3. Remember each dataset only has one graph that we need to access \n","  ##    (i.e. dataset[0])\n","\n","\n","  for dataset_name, dataset in datasets.items(): # loop over different datasets\n","    num_nodes = 0\n","    for label in dataset[0].node_label_index.keys(): # loop over different labels within a dataset\n","      num_nodes += dataset[0].node_label_index[label].shape[0]\n","    data_set_splits[dataset_name] = num_nodes\n","    \n","  #########################################\n","\n","  return data_set_splits\n","\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  dataset = GraphDataset([hete], task='node')\n","  # Splitting the dataset\n","  dataset_train, dataset_val, dataset_test = dataset.split(transductive=True, split_ratio=[0.4, 0.3, 0.3])\n","  datasets = {'train': dataset_train, 'val': dataset_val, 'test': dataset_test}\n","\n","  data_set_splits = compute_dataset_split_counts(datasets)\n","  for dataset_name, num_nodes in data_set_splits.items():\n","    print(\"{} dataset has {} nodes\".format(dataset_name, num_nodes))"]},{"cell_type":"markdown","metadata":{"id":"lFY2PaDbVKe4"},"source":["## DeepSNAP Dataset Visualization\n","\n","Now you can visualize the different nodes and edges used in each graph dataset split."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"elapsed":1067,"status":"ok","timestamp":1657143100456,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"iiyEw-agbgV8","outputId":"a94fe93e-55a9-4bdd-f2b0-69526f99424c"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 504x504 with 1 Axes>"],"image/png":"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\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["<Figure size 504x504 with 1 Axes>"],"image/png":"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\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["<Figure size 504x504 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["from deepsnap.dataset import GraphDataset\n","\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  dataset = GraphDataset([hete], task='node')\n","  # Splitting the dataset\n","  dataset_train, dataset_val, dataset_test = dataset.split(transductive=True, split_ratio=[0.4, 0.3, 0.3])\n","  titles = ['Train', 'Validation', 'Test']\n","\n","  for i, dataset in enumerate([dataset_train, dataset_val, dataset_test]):\n","    n0 = hete._convert_to_graph_index(dataset[0].node_label_index['n0'], 'n0').tolist()\n","    n1 = hete._convert_to_graph_index(dataset[0].node_label_index['n1'], 'n1').tolist()\n","\n","    plt.figure(figsize=(7, 7))\n","    plt.title(titles[i])\n","    nx.draw(G_orig, pos=pos, node_color=\"grey\", edge_color=colors, labels=labels, font_color='white')\n","    nx.draw_networkx_nodes(G_orig.subgraph(n0), pos=pos, node_color=\"blue\")\n","    nx.draw_networkx_nodes(G_orig.subgraph(n1), pos=pos, node_color=\"red\")\n","    show()"]},{"cell_type":"markdown","metadata":{"id":"e5LsVSRuI3hU"},"source":["# 2) Heterogeneous Graph Node Property Prediction\n","\n","Now that we have introduced you to the basics of Heterogeneous graphs, you will use PyTorch Geometric and DeepSNAP to implement a GNN model for heterogeneous graph node property prediction (node classification). You will draw upon your understanding of heterogeneous graphs from lecture and previous work in implementing GNN layers using PyG (introduced in Colab 3).\n","\n","First let's take a look at the general structure of a heterogeneous GNN layer by working through an example:\n","\n","Let's assume we have a graph $G$, which contains two node types $a$ and $b$, and three message types $m_1=(a, r_1, a)$, $m_2=(a, r_2, b)$ and $m_3=(a, r_3, b)$. For notation sake, we view **each message as (src, relation, dst)**, where messages \"flow\" from src to dst node types. \n","\n","When applying message passing in heterogenous graphs, we separately apply message passing over **each message type m separately**. For example, updating node type $b$ relies on two different message types $m_2$ and $m_3$. For the graph $G$, a heterogeneous GNN layer contains three separate Heterogeneous Message Passing layers (`HeteroGNNConv` in this Colab), where each `HeteroGNNConv` layer performs message passing and aggregation with respect to ***only one message type***. Since a message type is viewed as (src, relation, dst) and messages \"flow\" from src to dst, each `HeteroGNNConv` layer only computes embeddings for the *dst* nodes of a given message type. For example, the `HeteroGNNConv` layer for message type $m_2$ outputs updated embedding representations *only* for node's with type b. \n","\n","---\n","\n","An overview of the heterogeneous layer you will create is shown below:\n","\n","![test](https://drive.google.com/uc?export=view&id=1mkp4OeRrvC4iNFTXSywrmI6Pfl5J__gA)\n","\n","where we highlight the following notation:\n","\n","- $H_a^{(l)[m_1]}$ is the intermediate matrix of node embeddings for node type $a$, generated by the $l^{th}$ `HeteroGNNConv` layer for message type $m_1$.\n","- $H_a^{(l)}$ is the matrix with current embeddings for nodes of type $a$ after the $l^{th}$ layer of our Heterogeneous GNN model. Note that these embeddings can rely on one or more intermediate `HeteroGNNConv` layer embeddings(i.e. $H_b^{(l)}$ combines $H_b^{(l)[m_2]}$ and $H_b^{(l)[m_3]}$).\n","\n","Since each `HeteroGNNConv` is only applied over a single message type, we additionally define a Heterogeneous GNN Wrapper layer (`HeteroGNNWrapperConv`). This wrapper manages and combines the output of each `HeteroGNNConv` layer in order to generate the complete updated node embeddings for each node type in layer $l$ of our model. More specifically, the $l^{th}$ `HeteroGNNWrapperConv` layer takes as input the node embeddings computed for **each message type and node type** (e.g. $H_b^{(l)[m_2]}$ and $H_b^{(l)[m_3]}$) and **aggregates across message types** with the same $dst$ node type. The resulting output of the $l^{th}$ `HeteroGNNWrapperConv` layer is the updated embedding matrix $H_i^{(l)}$ for **each node type i**. \n","\n","Continuing on our example above, to compute the node embeddings $H_b^{(l)}$, the wrapper layer aggregates output embeddings from the `HeteroGNNConv` layers associated with message types $m_2$ and $m_3$ (i.e. $H_b^{(l)[m_2]}$ and $H_b^{(l)[m_3]}$). \n","\n","---\n","\n","With the `HeteroGNNWrapperConv` module, we can now draw a \"simplified\" heterogeneous layer structure as follows:\n","\n","<br/>\n","<center>\n","<img src=\"http://web.stanford.edu/class/cs224w/images/colab4/hetero_conv_1.png\"/>\n","</center>\n","<br/>\n","\n","---\n","**NOTE**: \n","As reference, it may be helpful to additionally read through PyG's introduction to heterogeneous graph representations and buidling heterogeneous GNN models: https://pytorch-geometric.readthedocs.io/en/latest/notes/heterogeneous.html "]},{"cell_type":"markdown","metadata":{"id":"rOTCyuRcJikS"},"source":["<font color='red'>Looking ahead, we recommend you implement the heterogeneous GNN model in following steps:</font>\n","\n","1. Implement `HeteroGNNConv`.\n","2. Implement **just** `mean` aggregation within `HeteroGNNWrapperConv`.\n","3. Implement `generate_convs`.\n","4. Implement the `HeteroGNN` model and the `train` function.\n","5. Train the model with `mean` aggregation and test your model to make sure your model has reasonable performance.\n","6. Once you are confident in your mean aggregation model, implement `attn` aggregation in `HeteroGNNWrapperConv`.\n","7. Train the model with `attn` aggregation and test your model to make sure your model has reasonable performance.\n","\n","Note: The key point of advice is to work completely through implementing the mean aggregation heterogeneous GNN model before diving into the more difficult attention based model."]},{"cell_type":"markdown","metadata":{"id":"pkFjcktiJJLm"},"source":["## Setup"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"NAm9_OcJJJ-W"},"outputs":[],"source":["import copy\n","import torch\n","import deepsnap\n","import random\n","import numpy as np\n","import torch.nn as nn\n","import torch.nn.functional as F\n","import torch_geometric.nn as pyg_nn\n","\n","from sklearn.metrics import f1_score\n","from deepsnap.hetero_gnn import forward_op\n","from deepsnap.hetero_graph import HeteroGraph\n","from torch_sparse import SparseTensor, matmul"]},{"cell_type":"markdown","metadata":{"id":"n2prITo3JSbo"},"source":["## Dataset\n","\n","You need to login to your Google account and enter the verification code below."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"QvQwxJX4JTJX"},"outputs":[],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  from pydrive.auth import GoogleAuth\n","  from pydrive.drive import GoogleDrive\n","  from google.colab import auth\n","  from oauth2client.client import GoogleCredentials\n","\n","  # Authenticate and create the PyDrive client\n","  auth.authenticate_user()\n","  gauth = GoogleAuth()\n","  gauth.credentials = GoogleCredentials.get_application_default()\n","  drive = GoogleDrive(gauth)"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"Igoy4F_xJbVn"},"outputs":[],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  id='1ivlxd6lJMcZ9taS44TMGG72x2V1GeVvk'\n","  downloaded = drive.CreateFile({'id': id})\n","  downloaded.GetContentFile('acm.pkl')"]},{"cell_type":"markdown","metadata":{"id":"VBlboS5kJmJL"},"source":["## Implementing `HeteroGNNConv`\n","\n","Now let's start working on your own implementation of the heterogeneous message passing layer (`HeteroGNNConv`)! Just as in Colab 3, you will implement the layer using PyTorch Geometric. \n","\n","At a high level, the `HeteroGNNConv` layer is equivalent to the homogenous GNN layers you implemented in Colab 3, but now **applied to an individual heterogeneous message type m**. Moreover, our heterogeneous GNN layer draws directly from the **GraphSAGE** message passing model ([Hamilton et al. (2017)](https://arxiv.org/abs/1706.02216)).\n","\n","You will begin by defining the `HeteroGNNConv` layer with respect to message type $m$:\n","\n","\\begin{equation}\n","m =(s, r, d)\n","\\end{equation}\n","\n","where each message type is a tuple containing three elements: $s$ - the source node type, $r$ - the edge (relation) type, and $d$ - the destination node type. \n","\n","The message passing update rule that you will implement is very similar to that of GraphSAGE, except you now need to **include the node types and the edge relation type**. The update rule for message type $m$ is described below:\n","\n","\\begin{equation}\n","h_v^{(l)[m]} = W^{(l)[m]} \\cdot \\text{CONCAT} \\Big( W_d^{(l)[m]} \\cdot h_v^{(l-1)}, W_s^{(l)[m]} \\cdot AGG(\\{h_u^{(l-1)}, \\forall u \\in N_{m}(v) \\})\\Big)\n","\\end{equation}\n","\n","where you compute $h_v^{(l)[m]}$, the node embedding representation for node $v$ after `HeteroGNNConv` layer $l$ with respect message type $m$. Further unpacking the formula:\n","- $W_s^{(l)[m]}$ - linear transformation matrix for the messages of neighboring source nodes of type $s$ along message type $m$.\n","- $W_d^{(l)[m]}$ - linear transformation matrix for the message from the node $v$ itself of type $d$.\n","- $W^{(l)[m]}$ - linear transformation matrix for the concatenated messages from neighboring node's and the central node.\n","- $h_u^{(l-1)}$ - the hidden embedding representation for node $u$ after the $(l-1)^{th}$ `HeteroGNNWrapperConv` layer. Note, that this embedding is not associated with a particular message type (see layer diagrams above). \n","- $N_{m}(v)$ - the set of neighbor source nodes $s$ for the node v that we are embedding along message type $m = (s, r, d)$. \n","\n","**NOTE**: We emphasize that each weight matrix is associated with a specific message type $[m]$ and additionally, the weight matrices applied to node messages are differentiated by node type (i.e. $W_s$ and $W_d$).\n","\n","Lastly, for simplicity, we use mean aggregations for $AGG$ where:\n","\n","\\begin{equation}\n","AGG(\\{h_u^{(l-1)}, \\forall u \\in N_{m}(v) \\}) = \\frac{1}{|N_{m}(v)|} \\sum_{u\\in N_{m}(v)} h_u^{(l-1)}\n","\\end{equation}"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"_Z1b0Mf8Jova"},"outputs":[],"source":["class HeteroGNNConv(pyg_nn.MessagePassing):\n","    def __init__(self, in_channels_src, in_channels_dst, out_channels):\n","        super(HeteroGNNConv, self).__init__(aggr=\"mean\")\n","\n","        self.in_channels_src = in_channels_src\n","        self.in_channels_dst = in_channels_dst\n","        self.out_channels = out_channels\n","\n","        # To simplify implementation, please initialize both self.lin_dst\n","        # and self.lin_src out_features to out_channels\n","        self.lin_dst = None\n","        self.lin_src = None\n","\n","        self.lin_update = None\n","\n","        ############# Your code here #############\n","        ## (~3 lines of code)\n","        ## Note:\n","        ## 1. Initialize the 3 linear layers.\n","        ## 2. Think through the connection between the mathematical\n","        ##    definition of the update rule and torch linear layers!\n","        self.lin_dst = torch.nn.Linear(in_features=self.in_channels_dst, out_features=self.out_channels)\n","        self.lin_src = torch.nn.Linear(in_features=self.in_channels_src, out_features=self.out_channels)\n","        self.lin_update = torch.nn.Linear(in_features=self.out_channels*2, out_features=self.out_channels) #(1,2*F') -> (1,F')\n","        ##########################################\n","\n","    def forward(\n","        self,\n","        node_feature_src,\n","        node_feature_dst,\n","        edge_index,\n","        size=None\n","    ):\n","        ############# Your code here #############\n","        ## (~1 line of code)\n","        ## Note:\n","        ## 1. Unlike Colab 3, we just need to call self.propagate with \n","        ## proper arguments.\n","        aggr_out = self.propagate(edge_index, size, node_feature_src=node_feature_src, node_feature_dst=node_feature_dst)\n","        return aggr_out\n","        ##########################################\n","\n","    def message_and_aggregate(self, edge_index, node_feature_src):\n","\n","        ############# Your code here #############\n","        ## (~1 line of code)\n","        ## Note:\n","        ## 1. Different from what we implemented in Colab 3, we use message_and_aggregate\n","        ##    to combine the previously seperate message and aggregate functions. \n","        ##    The benefit is that we can avoid materializing x_i and x_j\n","        ##    to make the implementation more efficient.\n","        ## 2. To implement efficiently, refer to PyG documentation for message_and_aggregate\n","        ##    and sparse-matrix multiplication:\n","        ##    https://pytorch-geometric.readthedocs.io/en/latest/notes/sparse_tensor.html\n","        ## 3. Here edge_index is torch_sparse SparseTensor. Although interesting, you\n","        ##    do not need to deeply understand SparseTensor represenations!\n","        ## 4. Conceptually, think through how the message passing and aggregation\n","        ##    expressed mathematically can be expressed through matrix multiplication.\n","        out = matmul(edge_index, node_feature_src, reduce=self.aggr)\n","        ##########################################\n","\n","        return out\n","\n","    def update(self, aggr_out, node_feature_dst):\n","\n","        ############# Your code here #############\n","        ## (~4 lines of code)\n","        ## Note:\n","        ## 1. The update function is called after message_and_aggregate\n","        ## 2. Think through the one-one connection between the mathematical update\n","        ##    rule and the 3 linear layers defined in the constructor. \n","        node_feature_dst_lin = self.lin_dst(node_feature_dst)\n","        aggr_out_lin = self.lin_src(aggr_out)\n","        node_feature_lin = torch.cat((node_feature_dst_lin, aggr_out_lin), dim=-1)\n","        aggr_out = self.lin_update(node_feature_lin)\n","        ##########################################\n","\n","        return aggr_out"]},{"cell_type":"markdown","metadata":{"id":"aKq8ScTiJthn"},"source":["## Heterogeneous GNN Wrapper Layer\n","\n","After implementing the `HeteroGNNConv` layer for each message type, you need to manage and aggregate the node embedding results (with respect to each message types). Here you will implement two types of **message type level aggregation**.\n","\n","The first one is simply **mean aggregation over message types**:\n","\n","\\begin{equation}\n","h_v^{(l)} = \\frac{1}{M}\\sum_{m=1}^{M}h_v^{(l)[m]}\n","\\end{equation}\n","\n","where node $v$ has node type $d$ and we sum over the $M$ message types that have destination node type $d$. From our original example, for a node v of type $b$ you aggregate v's `HeteroGNNConv` embeddings for message types $m_2$ and $m_3$ (i.e. $h_v^{(l)[m_2]}$ and $h_v^{(l)[m_3]}$).\n","\n","The second method you will implement is the semantic level attention introduced in **HAN** ([Wang et al. (2019)](https://arxiv.org/abs/1903.07293)). Instead of directly averaging on the message type aggregation results, you **use attention to learn which message type result is more important, then aggregate across all the message types**. Below are the equations for semantic level attention:\n","\n","\\begin{equation}\n","e_{m} = \\frac{1}{|V_{d}|} \\sum_{v \\in V_{d}} q_{attn}^{(l)T} \\cdot tanh \\Big( W_{attn}^{(l)} \\cdot h_v^{(l)[m]} + b \\Big)\n","\\end{equation}\n","\n","where $m$ is the message type and $d$ refers to the destination node type for that message ($m = (s, r, d)$). Additionally, $V_{d}$ refers to the set of nodes v with type $d$. Lastly, the unnormalized attention weight $e_m$ is a scaler computed for each message type $m$. \n","\n","Next, you can compute the normalized attention weights $\\alpha_{m}$ and update $h_v^{(l)}$:\n","\n","\\begin{equation}\n","\\alpha_{m} = \\frac{\\exp(e_{m})}{\\sum_{m=1}^M \\exp(e_{m})}\n","\\end{equation}\n","\n","\\begin{equation}\n","h_v^{(l)} = \\sum_{m=1}^{M} \\alpha_{m} \\cdot h_v^{(l)[m]}\n","\\end{equation}\n","\n",", where we emphasize that $M$ here is the number of message types associated with the destination node type $d$. \n","\n","**Note**: The implementation of the attention aggregation is tricky and nuanced. We strongly recommend working carefully through the math equations to undersatnd exactly what each notation refers to and how all the pieces fit together. If you can, try to connect the math to our original example, focusing on node type $b$, which depends on two different message types!"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"0_bun02xJwFm"},"outputs":[],"source":["class HeteroGNNWrapperConv(deepsnap.hetero_gnn.HeteroConv):\n","    def __init__(self, convs, args, aggr=None):\n","        super(HeteroGNNWrapperConv, self).__init__(convs, None)\n","        self.aggr = aggr\n","\n","        # Map the index and message type\n","        self.mapping = {}\n","\n","        # A numpy array that stores the final attention probability\n","        self.alpha = None\n","\n","        self.attn_proj = None\n","\n","        if self.aggr == \"attn\":\n","            ############# Your code here #############\n","            ## (~1 line of code)\n","            ## Note:\n","            ## 1. Initialize self.attn_proj, where self.attn_proj should include\n","            ##    two linear layers. Note, make sure you understand\n","            ##    which part of the equation self.attn_proj captures.\n","            ## 2. You should use nn.Sequential for self.attn_proj\n","            ## 3. nn.Linear and nn.Tanh are useful.\n","            ## 4. You can model a weight vector (rather than matrix) by using:\n","            ##    nn.Linear(some_size, 1, bias=False).\n","            ## 5. The first linear layer should have out_features as args['attn_size']\n","            ## 6. You can assume we only have one \"head\" for the attention.\n","            ## 7. We recommend you to implement the mean aggregation first. After \n","            ##    the mean aggregation works well in the training, then you can \n","            ##    implement this part.\n","\n","            # e_m\n","            self.attn_proj = nn.Sequential(\n","                nn.Linear(in_features=args['hidden_size'], out_features=args['attn_size'], bias=True),\n","                nn.Tanh(),\n","                nn.Linear(in_features=args['attn_size'], out_features=1, bias=False)\n","            )\n","\n","            ##########################################\n","    \n","    def reset_parameters(self):\n","        super(HeteroGNNWrapperConv, self).reset_parameters()\n","        if self.aggr == \"attn\":\n","            for layer in self.attn_proj.children():\n","                layer.reset_parameters()\n","    \n","    def forward(self, node_features, edge_indices):\n","        message_type_emb = {}\n","        for message_key, message_type in edge_indices.items():\n","            src_type, edge_type, dst_type = message_key\n","            node_feature_src = node_features[src_type]\n","            node_feature_dst = node_features[dst_type]\n","            edge_index = edge_indices[message_key]\n","            message_type_emb[message_key] = (\n","                self.convs[message_key](\n","                    node_feature_src,\n","                    node_feature_dst,\n","                    edge_index,\n","                )\n","            )\n","        node_emb = {dst: [] for _, _, dst in message_type_emb.keys()}\n","        mapping = {}        \n","        for (src, edge_type, dst), item in message_type_emb.items():\n","            mapping[len(node_emb[dst])] = (src, edge_type, dst)\n","            node_emb[dst].append(item)\n","        self.mapping = mapping\n","        for node_type, embs in node_emb.items():\n","            if len(embs) == 1:\n","                node_emb[node_type] = embs[0]\n","            else:\n","                node_emb[node_type] = self.aggregate(embs)\n","        return node_emb\n","    \n","    def aggregate(self, xs):\n","        # TODO: Implement this function that aggregates all message type results for one node type.\n","        # Here, xs is a list of tensors (embeddings) with respect to message \n","        # type aggregation results.\n","        \n","        # Useful dimensions from xs - particularly for `attn` aggregation\n","        N = xs[0].shape[0] # Number of nodes for the given node type\n","        M = len(xs) # Number of message types for the given node type\n","        # xs[0].shape [3025,64]\n","\n","        if self.aggr == \"mean\":\n","\n","            ############# Your code here #############\n","            ## (~2 lines of code)\n","            ## Note:\n","            ## 1. Explore the function parameter `xs`! \n","            aggr_out = torch.mean(torch.stack(xs), dim=0) # add dimension M\n","            return aggr_out\n","            ##########################################\n","\n","        elif self.aggr == \"attn\":\n","            ############# Your code here #############\n","            ## (~10 lines of code)\n","            ## Note:\n","            ## 1. Try to map out how the equations can be translated into code.\n","            ## 2. N and M defined above may be useful at least to understand.\n","            ## 3. Work first to compute the un-normalized attention weights e\n","            ##    for each message type - try to vectorize this!\n","            ## 4. torch.softmax and torch.cat are useful.\n","            ## 5. It might be useful to reshape and concatenate tensors using the  \n","            ##    `view()` function https://pytorch.org/docs/stable/tensor_view.html\n","            ##    and `torch.cat()`https://pytorch.org/docs/stable/generated/torch.cat.html\n","            ## 6. Store the value of attention alpha (as a numpy array) to self.alpha,\n","            ##    which has the shape (len(xs), ) i.e. (M,) attention weight self.alpha will be not be used \n","            ##    to backpropagate etc. in the model. We will use it to see how much \n","            ##    attention the layer pays on different message types.            \n","            \n","            # torch.stack(xs, dim=0) shape [2, 3025, 64] = [M, N, F]\n","            # e shape shape [2, 3025, 1]\n","            # e shape after [2, 1]\n","            e = self.attn_proj(torch.stack(xs, dim=0)) # linear transformation F->1\n","            e = torch.mean(e, dim=1) # average over number of nodes N\n","            # alpha shape [2,1]\n","            self.alpha = torch.softmax(e, dim=0) # softmax along dim of M\n","\n","            out = torch.stack(xs, dim=0)\n","            out = self.alpha.unsqueeze(-1) * out # [2,1,1] * [2,3025,64]\n","            out = torch.sum(out, dim=0) # sum along M dimension\n","\n","            self.alpha = self.alpha.squeeze().detach().cpu().numpy() #6. store self.alpha to numpy array\n","            return out \n","            ##########################################"]},{"cell_type":"markdown","metadata":{"id":"tn_pnCOKJw-d"},"source":["## Initialize Heterogeneous GNN Layers\n","\n","Now let's put it all together and initialize the Heterogeneous GNN Layers. Different from the homogeneous graph case, heterogeneous graphs can be a little bit complex.\n","\n","In general, you need to create a **dictionary** of `HeteroGNNConv` layers where the **keys are message types**.\n","\n","* To get all message types, `deepsnap.hetero_graph.HeteroGraph.message_types` is useful.\n","* When you initialize the first conv layers, you need to get the feature dimension of each node type. Using `deepsnap.hetero_graph.HeteroGraph.num_node_features(node_type)` will return the node feature dimension of `node_type`. In this function, you will set each `HeteroGNNConv` `out_channels` to be `hidden_size`.\n","* For the remaining conv layers, all node types will have the same embedding dimension `hidden_size` and we still set `HeteroGNNConv` `out_channels` to be `hidden_size` **for simplicity**.\n","\n"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"hSBImHClJzf4"},"outputs":[],"source":["def generate_convs(hetero_graph, conv, hidden_size, first_layer=False):\n","    # TODO: Implement this function that returns a dictionary of `HeteroGNNConv` \n","    # layers where the keys are message types. `hetero_graph` is deepsnap `HeteroGraph`\n","    # object and the `conv` is the `HeteroGNNConv`.\n","\n","    convs = {}\n","\n","    ############# Your code here #############\n","    ## (~9 lines of code)\n","    ## Note:\n","    ## 1. See the hints above!\n","    ## 2. conv is of type `HeteroGNNConv`\n","\n","    for message_type in hetero_graph.message_types: # iterate all message types in list\n","      if first_layer: \n","        # get node feaures dimention of src and dst nodes via message type (src, edge, dst)\n","        in_channels_src = hetero_graph.num_node_features(message_type[0])\n","        in_channels_dst = hetero_graph.num_node_features(message_type[-1])\n","        # first layer\n","        convs[message_type] = conv(in_channels_src, in_channels_dst, out_channels=hidden_size)\n","      else: # remaining conv layers\n","        convs[message_type] = conv(in_channels_src=hidden_size, in_channels_dst=hidden_size, out_channels=hidden_size)\n","    ##########################################\n","    \n","    return convs"]},{"cell_type":"markdown","metadata":{"id":"U39dX8EpJ3FG"},"source":["## HeteroGNN\n","\n","Now you will make a simple HeteroGNN model which contains only two `HeteroGNNWrapperConv` layers.\n","\n","For the forward function in `HeteroGNN`, the model is going to be run as following:\n","\n","$\\text{self.convs1} \\rightarrow \\text{self.bns1} \\rightarrow \\text{self.relus1} \\rightarrow \\text{self.convs2} \\rightarrow \\text{self.bns2} \\rightarrow \\text{self.relus2} \\rightarrow \\text{self.post_mps}$"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"rplknA8aJ6J5"},"outputs":[],"source":["class HeteroGNN(torch.nn.Module):\n","    def __init__(self, hetero_graph, args, aggr=\"mean\"):\n","        super(HeteroGNN, self).__init__()\n","\n","        self.aggr = aggr\n","        self.hidden_size = args['hidden_size']\n","\n","        self.convs1 = None\n","        self.convs2 = None\n","\n","        self.bns1 = nn.ModuleDict()\n","        self.bns2 = nn.ModuleDict()\n","        self.relus1 = nn.ModuleDict()\n","        self.relus2 = nn.ModuleDict()\n","        self.post_mps = nn.ModuleDict()\n","\n","        ############# Your code here #############\n","        ## (~10 lines of code)\n","        ## Note:\n","        ## 1. For self.convs1 and self.convs2, call generate_convs at first and then\n","        ##    pass the returned dictionary of `HeteroGNNConv` to `HeteroGNNWrapperConv`.\n","        ## 2. For self.bns, self.relus and self.post_mps, the keys are node_types.\n","        ##    `deepsnap.hetero_graph.HeteroGraph.node_types` will be helpful.\n","        ## 3. Initialize all batchnorms to torch.nn.BatchNorm1d(hidden_size, eps=1).\n","        ## 4. Initialize all relus to nn.LeakyReLU().\n","        ## 5. For self.post_mps, each value in the ModuleDict is a linear layer \n","        ##    where the `out_features` is the number of classes for that node type.\n","        ##    `deepsnap.hetero_graph.HeteroGraph.num_node_labels(node_type)` will be\n","        ##    useful.\n","\n","        # 1. generate convs for each message types and message type level aggregation\n","        convs1 = generate_convs(hetero_graph, conv=HeteroGNNConv, hidden_size=self.hidden_size, first_layer=True)\n","        self.convs1 = HeteroGNNWrapperConv(convs1, args, aggr)\n","        convs2 = generate_convs(hetero_graph, conv=HeteroGNNConv, hidden_size=self.hidden_size, first_layer=False)\n","        self.convs2 = HeteroGNNWrapperConv(convs2, args, aggr)\n","        # 2-5\n","        for node_type in hetero_graph.node_types:\n","          self.bns1[node_type] = torch.nn.BatchNorm1d(num_features=self.hidden_size, eps=1)\n","          self.bns2[node_type] = torch.nn.BatchNorm1d(num_features=self.hidden_size, eps=1)\n","          self.relus1[node_type] = torch.nn.LeakyReLU()\n","          self.relus2[node_type] = torch.nn.LeakyReLU()\n","          self.post_mps[node_type] = torch.nn.Linear(in_features=self.hidden_size, out_features=hetero_graph.num_node_labels(node_type))\n","         \n","        ##########################################\n","\n","    def forward(self, node_feature, edge_index):\n","        # TODO: Implement the forward function. Notice that `node_feature` is \n","        # a dictionary of tensors where keys are node types and values are \n","        # corresponding feature tensors. The `edge_index` is a dictionary of \n","        # tensors where keys are message types and values are corresponding\n","        # edge index tensors (with respect to each message type).\n","\n","        x = node_feature\n","\n","        ############# Your code here #############\n","        ## (~7 lines of code)\n","        ## Note:\n","        ## 1. `deepsnap.hetero_gnn.forward_op` can be helpful for\n","        ##    the bn, relu, and post_mp ops.\n","        x = self.convs1(x, edge_index)\n","        x = forward_op(x, self.bns1) # apply forward pass of the module_dict layer to x\n","        x = forward_op(x, self.relus1)\n","\n","        x = self.convs2(x, edge_index)\n","        x = forward_op(x, self.bns2)\n","        x = forward_op(x, self.relus2)\n","\n","        x = forward_op(x, self.post_mps)\n","        ##########################################\n","        \n","        return x\n","\n","    def loss(self, preds, y, indices):\n","        \n","        loss = 0\n","        loss_func = F.cross_entropy\n","\n","        ############# Your code here #############\n","        ## (~3 lines of code)\n","        ## Note:\n","        ## 1. For each node type in preds, accumulate computed loss to `loss`\n","        ## 2. Loss need to be computed with respect to the given index\n","        ## 3. preds is a dictionary of model predictions keyed by node_type.\n","        ## 4. indeces is a dictionary of labeled supervision nodes keyed\n","        ##    by node_type (train_idx)\n","        for node_type in preds.keys():\n","          idx = indices[node_type] # get value of index\n","          loss += loss_func(preds[node_type][idx], y[node_type][idx])\n","        ##########################################\n","\n","        return loss"]},{"cell_type":"markdown","metadata":{"id":"o9e7q_hUJ8zB"},"source":["## Training and Testing\n","\n","Here we provide you with the functions to train and test. You only need to implement one line of code here.\n","\n","**Please do not modify other parts in `train` and `test` for grading purposes.**"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"CI5Hl_5TJ_YL"},"outputs":[],"source":["import pandas as pd\n","\n","def train(model, optimizer, hetero_graph, train_idx):\n","    model.train()\n","    optimizer.zero_grad()\n","    preds = model(hetero_graph.node_feature, hetero_graph.edge_index)\n","\n","    loss = None\n","\n","    ############# Your code here #############\n","    ## Note:\n","    ## 1. Compute the loss here refering loss function defination above\n","    ## 2. `deepsnap.hetero_graph.HeteroGraph.node_label` is useful\n","    loss = model.loss(preds, hetero_graph.node_label, train_idx)\n","    ##########################################\n","\n","    loss.backward()\n","    optimizer.step()\n","    return loss.item()\n","\n","def test(model, graph, indices, best_model=None, best_val=0, save_preds=False, agg_type=None):\n","    model.eval()\n","    accs = []\n","    for i, index in enumerate(indices):\n","        preds = model(graph.node_feature, graph.edge_index)\n","        num_node_types = 0\n","        micro = 0\n","        macro = 0\n","        for node_type in preds:\n","            idx = index[node_type]\n","            pred = preds[node_type][idx]\n","            pred = pred.max(1)[1]\n","            label_np = graph.node_label[node_type][idx].cpu().numpy()\n","            pred_np = pred.cpu().numpy()\n","            micro = f1_score(label_np, pred_np, average='micro')\n","            macro = f1_score(label_np, pred_np, average='macro')\n","            num_node_types += 1\n","                  \n","        # Averaging f1 score might not make sense, but in our example we only\n","        # have one node type\n","        micro /= num_node_types\n","        macro /= num_node_types\n","        accs.append((micro, macro))\n","\n","        # Only save the test set predictions and labels!\n","        if save_preds and i == 2:\n","          print (\"Saving Heterogeneous Node Prediction Model Predictions with Agg:\", agg_type)\n","          print()\n","\n","          data = {}\n","          data['pred'] = pred_np\n","          data['label'] = label_np\n","\n","          df = pd.DataFrame(data=data)\n","          # Save locally as csv\n","          df.to_csv('ACM-Node-' + agg_type + 'Agg.csv', sep=',', index=False)\n","\n","    if accs[1][0] > best_val:\n","        best_val = accs[1][0]\n","        best_model = copy.deepcopy(model)\n","    return accs, best_model, best_val"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"DpNz9B5AKBUU"},"outputs":[],"source":["# Please do not change the following parameters\n","args = {\n","    'device': torch.device('cuda' if torch.cuda.is_available() else 'cpu'),\n","    'hidden_size': 64,\n","    'epochs': 100,\n","    'weight_decay': 1e-5,\n","    'lr': 0.003,\n","    'attn_size': 32,\n","}"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"x7POWyQqmAtH"},"outputs":[],"source":["def set_seed(seed=224):\n","\n","    random.seed(seed)\n","    np.random.seed(seed)\n","    torch.manual_seed(seed)\n","    torch.backends.cudnn.benchmark = False"]},{"cell_type":"markdown","metadata":{"id":"pRHbWD4hKED8"},"source":["## Dataset and Preprocessing\n","\n","Before testing out your model, you need to load the data and create a tensor backend (without a NetworkX graph) `deepsnap.hetero_graph.HeteroGraph` object.\n","\n","You will use the `ACM(3025)` dataset to test your model on a node property prediction task, which is proposed in **HAN** ([Wang et al. (2019)](https://arxiv.org/abs/1903.07293)) and extracted from [DGL](https://www.dgl.ai/)'s [ACM.mat](https://data.dgl.ai/dataset/ACM.mat).\n","\n","The original ACM dataset has three node types and two edge (relation) types. For simplicity, we simplify the heterogeneous graph to one node type and two edge types (shown below). This means that in you heterogeneous graph, we have one node type (paper) and two message types (paper, author, paper) and (paper, subject, paper).\n","\n","<br/>\n","<center>\n","<img src=\"http://web.stanford.edu/class/cs224w/images/colab4/cs224w-acm.png\"/>\n","</center>"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":4753,"status":"ok","timestamp":1657143109350,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"lJy03_IsKGh6","outputId":"6e2a66b6-0766-4959-b7f8-7083c8b1cc61"},"outputs":[{"output_type":"stream","name":"stdout","text":["Device: cuda\n","ACM heterogeneous graph: {'paper': 3025} nodes, {('paper', 'author', 'paper'): 26256, ('paper', 'subject', 'paper'): 2207736} edges\n","SparseTensor(row=tensor([   0,    0,    0,  ..., 3024, 3024, 3024], device='cuda:0'),\n","             col=tensor([   8,   20,   51,  ..., 2948, 2983, 2991], device='cuda:0'),\n","             size=(3025, 3025), nnz=26256, density=0.29%)\n","SparseTensor(row=tensor([   0,    0,    0,  ..., 3024, 3024, 3024], device='cuda:0'),\n","             col=tensor([  75,  434,  534,  ..., 3020, 3021, 3022], device='cuda:0'),\n","             size=(3025, 3025), nnz=2207736, density=24.13%)\n"]}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  print(\"Device: {}\".format(args['device']))\n","\n","  set_seed()\n","\n","  # Load the data\n","  data = torch.load(\"acm.pkl\")\n","\n","  # Message types\n","  message_type_1 = (\"paper\", \"author\", \"paper\")\n","  message_type_2 = (\"paper\", \"subject\", \"paper\")\n","\n","  # Dictionary of edge indices\n","  edge_index = {}\n","  edge_index[message_type_1] = data['pap']\n","  edge_index[message_type_2] = data['psp']\n","\n","  # Dictionary of node features\n","  node_feature = {}\n","  node_feature[\"paper\"] = data['feature']\n","\n","  # Dictionary of node labels\n","  node_label = {}\n","  node_label[\"paper\"] = data['label']\n","\n","  # Load the train, validation and test indices\n","  train_idx = {\"paper\": data['train_idx'].to(args['device'])}\n","  val_idx = {\"paper\": data['val_idx'].to(args['device'])}\n","  test_idx = {\"paper\": data['test_idx'].to(args['device'])}\n","\n","  # Construct a deepsnap tensor backend HeteroGraph\n","  hetero_graph = HeteroGraph(\n","      node_feature=node_feature,\n","      node_label=node_label,\n","      edge_index=edge_index,\n","      directed=True\n","  )\n","\n","  print(f\"ACM heterogeneous graph: {hetero_graph.num_nodes()} nodes, {hetero_graph.num_edges()} edges\")\n","\n","  # Node feature and node label to device\n","  for key in hetero_graph.node_feature:\n","      hetero_graph.node_feature[key] = hetero_graph.node_feature[key].to(args['device'])\n","  for key in hetero_graph.node_label:\n","      hetero_graph.node_label[key] = hetero_graph.node_label[key].to(args['device'])\n","\n","  # Edge_index to sparse tensor and to device\n","  for key in hetero_graph.edge_index:\n","      edge_index = hetero_graph.edge_index[key]\n","      adj = SparseTensor(row=edge_index[0], col=edge_index[1], sparse_sizes=(hetero_graph.num_nodes('paper'), hetero_graph.num_nodes('paper')))\n","      hetero_graph.edge_index[key] = adj.t().to(args['device'])\n","  print(hetero_graph.edge_index[message_type_1])\n","  print(hetero_graph.edge_index[message_type_2])"]},{"cell_type":"markdown","metadata":{"id":"hrmU5-QQKJv6"},"source":["## Start Training!\n","\n","Now let's start training!"]},{"cell_type":"markdown","metadata":{"id":"y0HplV9hKMkc"},"source":["## Training the Mean Aggregation"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":10267,"status":"ok","timestamp":1657143119614,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"ZgwfyzLbKOUw","outputId":"24ea540a-a8bc-4458-c437-b322675dba53"},"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1: loss 1.10178, train micro 33.33%, train macro 16.67%, valid micro 33.33%, valid macro 16.67%, test micro 32.89%, test macro 16.5%\n","Epoch 2: loss 1.09279, train micro 33.33%, train macro 16.67%, valid micro 33.33%, valid macro 16.67%, test micro 32.89%, test macro 16.5%\n","Epoch 3: loss 1.06362, train micro 85.33%, train macro 85.61%, valid micro 77.33%, valid macro 77.21%, test micro 63.34%, test macro 62.19%\n","Epoch 4: loss 1.00446, train micro 79.17%, train macro 77.13%, valid micro 76.33%, valid macro 73.73%, test micro 64.28%, test macro 58.73%\n","Epoch 5: loss 0.90438, train micro 73.33%, train macro 68.16%, valid micro 70.33%, valid macro 63.67%, test micro 66.21%, test macro 56.95%\n","Epoch 6: loss 0.76363, train micro 70.5%, train macro 63.09%, valid micro 67.67%, valid macro 58.37%, test micro 65.84%, test macro 55.33%\n","Epoch 7: loss 0.60617, train micro 70.17%, train macro 62.44%, valid micro 67.33%, valid macro 57.23%, test micro 65.74%, test macro 54.94%\n","Epoch 8: loss 0.46308, train micro 71.0%, train macro 63.97%, valid micro 68.0%, valid macro 58.7%, test micro 65.84%, test macro 54.98%\n","Epoch 9: loss 0.34652, train micro 72.17%, train macro 66.01%, valid micro 68.33%, valid macro 59.38%, test micro 66.07%, test macro 55.34%\n","Epoch 10: loss 0.25776, train micro 75.0%, train macro 70.58%, valid micro 69.67%, valid macro 61.98%, test micro 66.35%, test macro 55.83%\n","Epoch 11: loss 0.19302, train micro 78.5%, train macro 75.77%, valid micro 73.67%, valid macro 68.94%, test micro 66.73%, test macro 56.73%\n","Epoch 12: loss 0.14536, train micro 85.17%, train macro 84.13%, valid micro 81.67%, valid macro 79.98%, test micro 68.14%, test macro 59.61%\n","Epoch 13: loss 0.10947, train micro 92.17%, train macro 91.94%, valid micro 87.33%, valid macro 86.7%, test micro 70.45%, test macro 63.85%\n","Epoch 14: loss 0.08203, train micro 95.17%, train macro 95.1%, valid micro 89.67%, valid macro 89.23%, test micro 72.94%, test macro 67.97%\n","Epoch 15: loss 0.06161, train micro 96.67%, train macro 96.64%, valid micro 91.33%, valid macro 91.07%, test micro 75.11%, test macro 71.51%\n","Epoch 16: loss 0.04706, train micro 97.83%, train macro 97.83%, valid micro 92.0%, valid macro 91.81%, test micro 76.89%, test macro 74.2%\n","Epoch 17: loss 0.03685, train micro 98.5%, train macro 98.5%, valid micro 93.67%, valid macro 93.58%, test micro 78.54%, test macro 76.57%\n","Epoch 18: loss 0.02965, train micro 99.0%, train macro 99.0%, valid micro 95.0%, valid macro 94.97%, test micro 79.53%, test macro 77.96%\n","Epoch 19: loss 0.02425, train micro 99.33%, train macro 99.33%, valid micro 95.33%, valid macro 95.35%, test micro 80.05%, test macro 78.71%\n","Epoch 20: loss 0.02029, train micro 99.33%, train macro 99.33%, valid micro 95.33%, valid macro 95.35%, test micro 80.28%, test macro 79.07%\n","Epoch 21: loss 0.01727, train micro 99.67%, train macro 99.67%, valid micro 95.33%, valid macro 95.35%, test micro 80.89%, test macro 79.8%\n","Epoch 22: loss 0.01489, train micro 100.0%, train macro 100.0%, valid micro 95.33%, valid macro 95.35%, test micro 81.32%, test macro 80.24%\n","Epoch 23: loss 0.01297, train micro 100.0%, train macro 100.0%, valid micro 95.33%, valid macro 95.35%, test micro 81.22%, test macro 80.17%\n","Epoch 24: loss 0.01136, train micro 100.0%, train macro 100.0%, valid micro 95.33%, valid macro 95.35%, test micro 81.6%, test macro 80.55%\n","Epoch 25: loss 0.00984, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.12%, test macro 81.13%\n","Epoch 26: loss 0.00845, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 82.35%, test macro 81.38%\n","Epoch 27: loss 0.00728, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.66%, test micro 82.73%, test macro 81.8%\n","Epoch 28: loss 0.00631, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.99%, test micro 83.39%, test macro 82.57%\n","Epoch 29: loss 0.00558, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.53%, test macro 82.74%\n","Epoch 30: loss 0.005, train micro 100.0%, train macro 100.0%, valid micro 97.67%, valid macro 97.66%, test micro 83.72%, test macro 82.96%\n","Epoch 31: loss 0.00445, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.32%, test micro 83.67%, test macro 82.93%\n","Epoch 32: loss 0.00382, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.32%, test micro 83.76%, test macro 83.05%\n","Epoch 33: loss 0.00319, train micro 100.0%, train macro 100.0%, valid micro 98.0%, valid macro 97.99%, test micro 83.86%, test macro 83.18%\n","Epoch 34: loss 0.00264, train micro 100.0%, train macro 100.0%, valid micro 98.67%, valid macro 98.66%, test micro 83.76%, test macro 83.15%\n","Epoch 35: loss 0.00219, train micro 100.0%, train macro 100.0%, valid micro 98.33%, valid macro 98.33%, test micro 83.91%, test macro 83.31%\n","Epoch 36: loss 0.00183, train micro 100.0%, train macro 100.0%, valid micro 98.33%, valid macro 98.34%, test micro 83.72%, test macro 83.12%\n","Epoch 37: loss 0.00155, train micro 100.0%, train macro 100.0%, valid micro 98.33%, valid macro 98.34%, test micro 83.81%, test macro 83.23%\n","Epoch 38: loss 0.00134, train micro 100.0%, train macro 100.0%, valid micro 98.33%, valid macro 98.34%, test micro 83.91%, test macro 83.35%\n","Epoch 39: loss 0.00117, train micro 100.0%, train macro 100.0%, valid micro 98.0%, valid macro 98.0%, test micro 83.76%, test macro 83.21%\n","Epoch 40: loss 0.00105, train micro 100.0%, train macro 100.0%, valid micro 98.0%, valid macro 98.0%, test micro 83.62%, test macro 83.09%\n","Epoch 41: loss 0.00095, train micro 100.0%, train macro 100.0%, valid micro 98.0%, valid macro 98.0%, test micro 83.53%, test macro 83.01%\n","Epoch 42: loss 0.00087, train micro 100.0%, train macro 100.0%, valid micro 97.67%, valid macro 97.67%, test micro 83.53%, test macro 83.02%\n","Epoch 43: loss 0.00081, train micro 100.0%, train macro 100.0%, valid micro 97.67%, valid macro 97.67%, test micro 83.53%, test macro 83.03%\n","Epoch 44: loss 0.00076, train micro 100.0%, train macro 100.0%, valid micro 98.0%, valid macro 98.0%, test micro 83.48%, test macro 83.01%\n","Epoch 45: loss 0.00072, train micro 100.0%, train macro 100.0%, valid micro 97.67%, valid macro 97.66%, test micro 83.44%, test macro 82.97%\n","Epoch 46: loss 0.00068, train micro 100.0%, train macro 100.0%, valid micro 97.67%, valid macro 97.66%, test micro 83.48%, test macro 83.03%\n","Epoch 47: loss 0.00065, train micro 100.0%, train macro 100.0%, valid micro 97.67%, valid macro 97.66%, test micro 83.67%, test macro 83.27%\n","Epoch 48: loss 0.00062, train micro 100.0%, train macro 100.0%, valid micro 97.67%, valid macro 97.66%, test micro 83.76%, test macro 83.37%\n","Epoch 49: loss 0.00059, train micro 100.0%, train macro 100.0%, valid micro 97.67%, valid macro 97.66%, test micro 83.62%, test macro 83.24%\n","Epoch 50: loss 0.00057, train micro 100.0%, train macro 100.0%, valid micro 97.67%, valid macro 97.66%, test micro 83.58%, test macro 83.22%\n","Epoch 51: loss 0.00055, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.58%, test macro 83.25%\n","Epoch 52: loss 0.00053, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.62%, test macro 83.31%\n","Epoch 53: loss 0.00051, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.72%, test macro 83.43%\n","Epoch 54: loss 0.0005, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.76%, test macro 83.51%\n","Epoch 55: loss 0.00048, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.76%, test macro 83.51%\n","Epoch 56: loss 0.00047, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.72%, test macro 83.47%\n","Epoch 57: loss 0.00046, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.62%, test macro 83.39%\n","Epoch 58: loss 0.00044, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.67%, test macro 83.45%\n","Epoch 59: loss 0.00043, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.48%, test macro 83.28%\n","Epoch 60: loss 0.00042, train micro 100.0%, train macro 100.0%, valid micro 97.33%, valid macro 97.33%, test micro 83.39%, test macro 83.18%\n","Epoch 61: loss 0.00041, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.99%, test micro 83.34%, test macro 83.14%\n","Epoch 62: loss 0.00041, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.99%, test micro 83.11%, test macro 82.92%\n","Epoch 63: loss 0.0004, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.65%, test micro 83.06%, test macro 82.88%\n","Epoch 64: loss 0.00039, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 83.06%, test macro 82.89%\n","Epoch 65: loss 0.00038, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 82.96%, test macro 82.8%\n","Epoch 66: loss 0.00038, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 82.96%, test macro 82.8%\n","Epoch 67: loss 0.00037, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 82.92%, test macro 82.76%\n","Epoch 68: loss 0.00036, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 82.87%, test macro 82.71%\n","Epoch 69: loss 0.00036, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 82.87%, test macro 82.71%\n","Epoch 70: loss 0.00035, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 82.92%, test macro 82.76%\n","Epoch 71: loss 0.00035, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 83.01%, test macro 82.87%\n","Epoch 72: loss 0.00034, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 83.06%, test macro 82.91%\n","Epoch 73: loss 0.00034, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 83.11%, test macro 82.96%\n","Epoch 74: loss 0.00033, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 83.11%, test macro 82.97%\n","Epoch 75: loss 0.00033, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.98%, test micro 83.11%, test macro 82.97%\n","Epoch 76: loss 0.00032, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.15%, test macro 83.02%\n","Epoch 77: loss 0.00032, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.2%, test macro 83.08%\n","Epoch 78: loss 0.00032, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.29%, test macro 83.18%\n","Epoch 79: loss 0.00031, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.25%, test macro 83.13%\n","Epoch 80: loss 0.00031, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.25%, test macro 83.13%\n","Epoch 81: loss 0.0003, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.25%, test macro 83.13%\n","Epoch 82: loss 0.0003, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.15%, test macro 83.03%\n","Epoch 83: loss 0.0003, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.15%, test macro 83.03%\n","Epoch 84: loss 0.0003, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.15%, test macro 83.03%\n","Epoch 85: loss 0.00029, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.15%, test macro 83.03%\n","Epoch 86: loss 0.00029, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.15%, test macro 83.03%\n","Epoch 87: loss 0.00029, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.15%, test macro 83.03%\n","Epoch 88: loss 0.00028, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.64%, test micro 83.15%, test macro 83.03%\n","Epoch 89: loss 0.00028, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.15%, test macro 83.03%\n","Epoch 90: loss 0.00028, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.2%, test macro 83.08%\n","Epoch 91: loss 0.00028, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.2%, test macro 83.08%\n","Epoch 92: loss 0.00027, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.15%, test macro 83.04%\n","Epoch 93: loss 0.00027, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.15%, test macro 83.04%\n","Epoch 94: loss 0.00027, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.11%, test macro 82.99%\n","Epoch 95: loss 0.00027, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.11%, test macro 82.99%\n","Epoch 96: loss 0.00026, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.11%, test macro 82.99%\n","Epoch 97: loss 0.00026, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.11%, test macro 82.99%\n","Epoch 98: loss 0.00026, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.01%, test macro 82.89%\n","Epoch 99: loss 0.00026, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.01%, test macro 82.9%\n","Epoch 100: loss 0.00026, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 96.97%, test micro 83.01%, test macro 82.9%\n","Saving Heterogeneous Node Prediction Model Predictions with Agg: Mean\n","\n","Best model: train micro 100.0%, train macro 100.0%, valid micro 98.67%, valid macro 98.66%, test micro 83.76%, test macro 83.15%\n"]}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  best_model = None\n","  best_val = 0\n","\n","  set_seed()\n","\n","  model = HeteroGNN(hetero_graph, args, aggr=\"mean\").to(args['device'])\n","  optimizer = torch.optim.Adam(model.parameters(), lr=args['lr'], weight_decay=args['weight_decay'])\n","\n","  for epoch in range(args['epochs']):\n","      loss = train(model, optimizer, hetero_graph, train_idx)\n","      accs, best_model, best_val = test(model, hetero_graph, [train_idx, val_idx, test_idx], best_model, best_val)\n","      print(\n","          f\"Epoch {epoch + 1}: loss {round(loss, 5)}, \"\n","          f\"train micro {round(accs[0][0] * 100, 2)}%, train macro {round(accs[0][1] * 100, 2)}%, \"\n","          f\"valid micro {round(accs[1][0] * 100, 2)}%, valid macro {round(accs[1][1] * 100, 2)}%, \"\n","          f\"test micro {round(accs[2][0] * 100, 2)}%, test macro {round(accs[2][1] * 100, 2)}%\"\n","      )\n","  best_accs, _, _ = test(best_model, hetero_graph, [train_idx, val_idx, test_idx], save_preds=True, agg_type=\"Mean\")\n","  print(\n","      f\"Best model: \"\n","      f\"train micro {round(best_accs[0][0] * 100, 2)}%, train macro {round(best_accs[0][1] * 100, 2)}%, \"\n","      f\"valid micro {round(best_accs[1][0] * 100, 2)}%, valid macro {round(best_accs[1][1] * 100, 2)}%, \"\n","      f\"test micro {round(best_accs[2][0] * 100, 2)}%, test macro {round(best_accs[2][1] * 100, 2)}%\"\n","  )"]},{"cell_type":"markdown","metadata":{"id":"rtkKBI_nKS1T"},"source":["## Question 2.1: What is your maximum test set **micro** F1 score for the best_model when using mean aggregation? (10 points)\n"]},{"cell_type":"markdown","metadata":{"id":"AIvw51jMKTvn"},"source":["## Question 2.2: What is your maximum test set **macro** F1 score for the best_model when using the mean aggregation? (10 points)\n"]},{"cell_type":"markdown","metadata":{"id":"JBiYvwcuKd0z"},"source":["## Training the Attention Aggregation"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":11822,"status":"ok","timestamp":1657143131432,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"e6na5zyQKfvi","outputId":"a14b5b81-6293-4941-85e0-7e5df0c9f632"},"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1: loss 1.09868, train micro 33.33%, train macro 16.67%, valid micro 33.33%, valid macro 16.67%, test micro 35.81%, test macro 17.58%\n","Epoch 2: loss 1.09267, train micro 35.5%, train macro 21.01%, valid micro 33.33%, valid macro 16.67%, test micro 35.91%, test macro 17.79%\n","Epoch 3: loss 1.07109, train micro 65.33%, train macro 54.49%, valid micro 63.33%, valid macro 53.08%, test micro 62.87%, test macro 52.8%\n","Epoch 4: loss 1.02665, train micro 66.33%, train macro 55.41%, valid micro 66.33%, valid macro 55.31%, test micro 65.08%, test macro 54.36%\n","Epoch 5: loss 0.94703, train micro 67.17%, train macro 56.77%, valid micro 66.33%, valid macro 55.22%, test micro 65.32%, test macro 54.46%\n","Epoch 6: loss 0.8225, train micro 68.0%, train macro 58.42%, valid micro 66.67%, valid macro 55.96%, test micro 65.41%, test macro 54.56%\n","Epoch 7: loss 0.6564, train micro 70.0%, train macro 62.32%, valid micro 67.33%, valid macro 57.39%, test micro 65.55%, test macro 54.68%\n","Epoch 8: loss 0.47672, train micro 72.5%, train macro 66.7%, valid micro 68.67%, valid macro 60.11%, test micro 65.74%, test macro 55.01%\n","Epoch 9: loss 0.32396, train micro 76.33%, train macro 72.7%, valid micro 70.0%, valid macro 62.59%, test micro 66.07%, test macro 55.72%\n","Epoch 10: loss 0.21813, train micro 82.33%, train macro 80.85%, valid micro 75.33%, valid macro 71.77%, test micro 66.92%, test macro 57.52%\n","Epoch 11: loss 0.15063, train micro 88.33%, train macro 87.83%, valid micro 80.67%, valid macro 79.09%, test micro 68.19%, test macro 60.18%\n","Epoch 12: loss 0.10621, train micro 93.33%, train macro 93.19%, valid micro 86.67%, valid macro 86.2%, test micro 70.49%, test macro 64.45%\n","Epoch 13: loss 0.07497, train micro 96.5%, train macro 96.47%, valid micro 90.0%, valid macro 89.82%, test micro 72.94%, test macro 68.28%\n","Epoch 14: loss 0.05362, train micro 98.17%, train macro 98.16%, valid micro 91.0%, valid macro 90.82%, test micro 75.86%, test macro 72.83%\n","Epoch 15: loss 0.03974, train micro 98.5%, train macro 98.5%, valid micro 92.67%, valid macro 92.56%, test micro 78.82%, test macro 76.8%\n","Epoch 16: loss 0.03061, train micro 98.83%, train macro 98.83%, valid micro 94.67%, valid macro 94.62%, test micro 80.75%, test macro 79.31%\n","Epoch 17: loss 0.02426, train micro 99.33%, train macro 99.33%, valid micro 96.33%, valid macro 96.31%, test micro 82.82%, test macro 81.84%\n","Epoch 18: loss 0.0197, train micro 99.83%, train macro 99.83%, valid micro 96.67%, valid macro 96.65%, test micro 83.86%, test macro 83.08%\n","Epoch 19: loss 0.01644, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.32%, test micro 84.14%, test macro 83.41%\n","Epoch 20: loss 0.01396, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.32%, test micro 84.09%, test macro 83.37%\n","Epoch 21: loss 0.01189, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.66%, test micro 83.67%, test macro 82.92%\n","Epoch 22: loss 0.01015, train micro 99.83%, train macro 99.83%, valid micro 96.0%, valid macro 95.99%, test micro 83.34%, test macro 82.54%\n","Epoch 23: loss 0.00871, train micro 99.83%, train macro 99.83%, valid micro 96.0%, valid macro 95.99%, test micro 82.82%, test macro 81.94%\n","Epoch 24: loss 0.00752, train micro 99.83%, train macro 99.83%, valid micro 95.67%, valid macro 95.64%, test micro 82.12%, test macro 81.1%\n","Epoch 25: loss 0.00656, train micro 99.83%, train macro 99.83%, valid micro 95.33%, valid macro 95.3%, test micro 81.6%, test macro 80.48%\n","Epoch 26: loss 0.00579, train micro 99.83%, train macro 99.83%, valid micro 95.33%, valid macro 95.3%, test micro 81.51%, test macro 80.4%\n","Epoch 27: loss 0.00515, train micro 99.83%, train macro 99.83%, valid micro 95.33%, valid macro 95.3%, test micro 81.51%, test macro 80.41%\n","Epoch 28: loss 0.00458, train micro 99.83%, train macro 99.83%, valid micro 95.0%, valid macro 94.96%, test micro 81.55%, test macro 80.5%\n","Epoch 29: loss 0.00402, train micro 100.0%, train macro 100.0%, valid micro 95.33%, valid macro 95.31%, test micro 81.55%, test macro 80.52%\n","Epoch 30: loss 0.0035, train micro 100.0%, train macro 100.0%, valid micro 95.67%, valid macro 95.64%, test micro 81.74%, test macro 80.78%\n","Epoch 31: loss 0.00304, train micro 100.0%, train macro 100.0%, valid micro 95.33%, valid macro 95.32%, test micro 82.16%, test macro 81.26%\n","Epoch 32: loss 0.00263, train micro 100.0%, train macro 100.0%, valid micro 95.67%, valid macro 95.66%, test micro 82.31%, test macro 81.4%\n","Epoch 33: loss 0.00228, train micro 100.0%, train macro 100.0%, valid micro 95.33%, valid macro 95.32%, test micro 82.78%, test macro 81.93%\n","Epoch 34: loss 0.00198, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 95.99%, test micro 82.54%, test macro 81.72%\n","Epoch 35: loss 0.00173, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 95.99%, test micro 82.64%, test macro 81.88%\n","Epoch 36: loss 0.00153, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 95.99%, test micro 82.64%, test macro 81.89%\n","Epoch 37: loss 0.00136, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 95.99%, test micro 82.96%, test macro 82.27%\n","Epoch 38: loss 0.00123, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 95.99%, test micro 83.2%, test macro 82.53%\n","Epoch 39: loss 0.00111, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 95.99%, test micro 83.25%, test macro 82.62%\n","Epoch 40: loss 0.00102, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.33%, test micro 83.06%, test macro 82.45%\n","Epoch 41: loss 0.00094, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.33%, test micro 83.11%, test macro 82.57%\n","Epoch 42: loss 0.00088, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.33%, test micro 83.06%, test macro 82.56%\n","Epoch 43: loss 0.00082, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.44%, test macro 82.99%\n","Epoch 44: loss 0.00077, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.34%, test macro 82.91%\n","Epoch 45: loss 0.00073, train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 97.0%, test micro 83.29%, test macro 82.86%\n","Epoch 46: loss 0.00069, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.48%, test macro 83.1%\n","Epoch 47: loss 0.00066, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.34%, test macro 82.95%\n","Epoch 48: loss 0.00063, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.15%, test macro 82.78%\n","Epoch 49: loss 0.0006, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.2%, test macro 82.85%\n","Epoch 50: loss 0.00058, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.01%, test macro 82.66%\n","Epoch 51: loss 0.00055, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.11%, test macro 82.78%\n","Epoch 52: loss 0.00053, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.11%, test macro 82.8%\n","Epoch 53: loss 0.00052, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 83.11%, test macro 82.8%\n","Epoch 54: loss 0.0005, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.67%, test micro 82.96%, test macro 82.67%\n","Epoch 55: loss 0.00049, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 83.06%, test macro 82.77%\n","Epoch 56: loss 0.00047, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.92%, test macro 82.63%\n","Epoch 57: loss 0.00046, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.73%, test macro 82.44%\n","Epoch 58: loss 0.00045, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.78%, test macro 82.48%\n","Epoch 59: loss 0.00044, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.82%, test macro 82.52%\n","Epoch 60: loss 0.00043, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.78%, test macro 82.49%\n","Epoch 61: loss 0.00042, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.78%, test macro 82.5%\n","Epoch 62: loss 0.00041, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.73%, test macro 82.45%\n","Epoch 63: loss 0.0004, train micro 100.0%, train macro 100.0%, valid micro 95.67%, valid macro 95.68%, test micro 82.73%, test macro 82.45%\n","Epoch 64: loss 0.00039, train micro 100.0%, train macro 100.0%, valid micro 95.67%, valid macro 95.68%, test micro 82.64%, test macro 82.36%\n","Epoch 65: loss 0.00038, train micro 100.0%, train macro 100.0%, valid micro 95.67%, valid macro 95.68%, test micro 82.64%, test macro 82.36%\n","Epoch 66: loss 0.00037, train micro 100.0%, train macro 100.0%, valid micro 95.67%, valid macro 95.68%, test micro 82.49%, test macro 82.21%\n","Epoch 67: loss 0.00037, train micro 100.0%, train macro 100.0%, valid micro 95.67%, valid macro 95.68%, test micro 82.49%, test macro 82.22%\n","Epoch 68: loss 0.00036, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.54%, test macro 82.27%\n","Epoch 69: loss 0.00036, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.49%, test macro 82.22%\n","Epoch 70: loss 0.00035, train micro 100.0%, train macro 100.0%, valid micro 96.0%, valid macro 96.01%, test micro 82.49%, test macro 82.22%\n","Epoch 71: loss 0.00034, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.49%, test macro 82.22%\n","Epoch 72: loss 0.00034, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.49%, test macro 82.22%\n","Epoch 73: loss 0.00033, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.49%, test macro 82.22%\n","Epoch 74: loss 0.00033, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.54%, test macro 82.27%\n","Epoch 75: loss 0.00032, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.54%, test macro 82.27%\n","Epoch 76: loss 0.00032, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.54%, test macro 82.27%\n","Epoch 77: loss 0.00032, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.54%, test macro 82.27%\n","Epoch 78: loss 0.00031, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.54%, test macro 82.27%\n","Epoch 79: loss 0.00031, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.49%, test macro 82.22%\n","Epoch 80: loss 0.0003, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.45%, test macro 82.16%\n","Epoch 81: loss 0.0003, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.35%, test macro 82.08%\n","Epoch 82: loss 0.0003, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.31%, test macro 82.04%\n","Epoch 83: loss 0.00029, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.35%, test macro 82.09%\n","Epoch 84: loss 0.00029, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.35%, test macro 82.09%\n","Epoch 85: loss 0.00029, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.4%, test macro 82.14%\n","Epoch 86: loss 0.00028, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.26%, test macro 82.01%\n","Epoch 87: loss 0.00028, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.34%, test micro 82.21%, test macro 81.97%\n","Epoch 88: loss 0.00028, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.68%, test micro 82.26%, test macro 82.01%\n","Epoch 89: loss 0.00027, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.68%, test micro 82.26%, test macro 82.02%\n","Epoch 90: loss 0.00027, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.68%, test micro 82.26%, test macro 82.02%\n","Epoch 91: loss 0.00027, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.68%, test micro 82.26%, test macro 82.02%\n","Epoch 92: loss 0.00027, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.68%, test micro 82.26%, test macro 82.02%\n","Epoch 93: loss 0.00026, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.68%, test micro 82.26%, test macro 82.02%\n","Epoch 94: loss 0.00026, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.68%, test micro 82.21%, test macro 81.98%\n","Epoch 95: loss 0.00026, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.68%, test micro 82.12%, test macro 81.89%\n","Epoch 96: loss 0.00026, train micro 100.0%, train macro 100.0%, valid micro 96.67%, valid macro 96.68%, test micro 82.12%, test macro 81.89%\n","Epoch 97: loss 0.00025, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.35%, test micro 82.16%, test macro 81.94%\n","Epoch 98: loss 0.00025, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.35%, test micro 82.07%, test macro 81.85%\n","Epoch 99: loss 0.00025, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.35%, test micro 82.07%, test macro 81.86%\n","Epoch 100: loss 0.00025, train micro 100.0%, train macro 100.0%, valid micro 96.33%, valid macro 96.35%, test micro 82.07%, test macro 81.86%\n","Saving Heterogeneous Node Prediction Model Predictions with Agg: Attention\n","\n","Best model: train micro 100.0%, train macro 100.0%, valid micro 97.0%, valid macro 97.0%, test micro 83.29%, test macro 82.86%\n"]}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  best_model = None\n","  best_val = 0\n","\n","  set_seed()\n","\n","  output_size = hetero_graph.num_node_labels('paper')\n","  model = HeteroGNN(hetero_graph, args, aggr=\"attn\").to(args['device'])\n","  optimizer = torch.optim.Adam(model.parameters(), lr=args['lr'], weight_decay=args['weight_decay'])\n","\n","  for epoch in range(args['epochs']):\n","      loss = train(model, optimizer, hetero_graph, train_idx)\n","      accs, best_model, best_val = test(model, hetero_graph, [train_idx, val_idx, test_idx], best_model, best_val)\n","      print(\n","          f\"Epoch {epoch + 1}: loss {round(loss, 5)}, \"\n","          f\"train micro {round(accs[0][0] * 100, 2)}%, train macro {round(accs[0][1] * 100, 2)}%, \"\n","          f\"valid micro {round(accs[1][0] * 100, 2)}%, valid macro {round(accs[1][1] * 100, 2)}%, \"\n","          f\"test micro {round(accs[2][0] * 100, 2)}%, test macro {round(accs[2][1] * 100, 2)}%\"\n","      )\n","  best_accs, _, _ = test(best_model, hetero_graph, [train_idx, val_idx, test_idx], save_preds=True, agg_type=\"Attention\")\n","  print(\n","      f\"Best model: \"\n","      f\"train micro {round(best_accs[0][0] * 100, 2)}%, train macro {round(best_accs[0][1] * 100, 2)}%, \"\n","      f\"valid micro {round(best_accs[1][0] * 100, 2)}%, valid macro {round(best_accs[1][1] * 100, 2)}%, \"\n","      f\"test micro {round(best_accs[2][0] * 100, 2)}%, test macro {round(best_accs[2][1] * 100, 2)}%\"\n","  )"]},{"cell_type":"markdown","metadata":{"id":"mtAhFLQQKgbl"},"source":["## Question 2.3: What is your maximum test set **micro** F1 score for the best_model when using the attention aggregation? (10 points)\n"]},{"cell_type":"markdown","metadata":{"id":"_cnsMGbsqJG_"},"source":["## Question 2.4: What is your maximum test set **macro** F1 score for the best_model when using the attention aggregation? (10 points)\n"]},{"cell_type":"markdown","metadata":{"id":"aQgx5y4UqMHH"},"source":["## Attention for each Message Type\n","\n","Through **message type level attention**, you can learn which message type is more important to which layer.\n","\n","Here you can see how each layer pays different attention to each message type."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":24,"status":"ok","timestamp":1657143131432,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"hvK58gijqN_C","outputId":"660afe38-4dff-4185-a7a2-730940ebd90e"},"outputs":[{"output_type":"stream","name":"stdout","text":["Layer 1 has attention 0.26564013957977295 on message type ('paper', 'author', 'paper')\n","Layer 1 has attention 0.734359860420227 on message type ('paper', 'subject', 'paper')\n","Layer 2 has attention 0.48180606961250305 on message type ('paper', 'author', 'paper')\n","Layer 2 has attention 0.5181939601898193 on message type ('paper', 'subject', 'paper')\n"]}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  if model.convs1.alpha is not None and model.convs2.alpha is not None:\n","      for idx, message_type in model.convs1.mapping.items():\n","          print(f\"Layer 1 has attention {model.convs1.alpha[idx]} on message type {message_type}\")\n","      for idx, message_type in model.convs2.mapping.items():\n","          print(f\"Layer 2 has attention {model.convs2.alpha[idx]} on message type {message_type}\")"]},{"cell_type":"markdown","metadata":{"id":"SxkYLgxAOxz7"},"source":["# 3) Neighbor Sampling\n","\n","In this final part of the Colab, we shift gears slightly and give a preview into the work for Colab5. Here we introduce an advanced topic in GNNs, where you will work with PyTorch Geometric's `NeighborSampler` to scale up training and testing of GNN's on the OGB `arxiv` dataset. Neighbor Sampling, originally proposed in **GraphSAGE** ([Hamilton et al. (2017)](https://arxiv.org/abs/1706.02216)), is a representative method to **scale up GNNs**. As we learned in lecture, rather than loading the entire graph into memory for each training loop, you can instead sample a mini-batch of the nodes you want to embed and **only** load the K-hop graph neighborhoods needed to embed these nodes. In this way you take advantage of the fact that **the embedding of a node u only depends on its K-hop neighborhood. **To further reduce the memory footprint and computational cost, you can choose to sample only a subset of a node's neighborhood during message passing and aggregation."]},{"cell_type":"markdown","metadata":{"id":"Kho6SHUVO1ny"},"source":["## Setup"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"X1WJLGKsOx_k"},"outputs":[],"source":["import os\n","import copy\n","import torch\n","import pandas as pd\n","import torch.nn.functional as F\n","import torch_geometric.transforms as T\n","\n","from torch_geometric.nn import SAGEConv\n","from torch_geometric.loader import NeighborSampler\n","\n","# Install OGB\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  !pip install -q ogb\n","\n","from ogb.nodeproppred import PygNodePropPredDataset, Evaluator"]},{"cell_type":"markdown","metadata":{"id":"UKqZWqRbO7km"},"source":["## PyTorch Geometric Neighbor Sampler\n","\n","PyTorch Geometric has implemented Neighbor Samplinging through the [NeighborSampler](https://pytorch-geometric.readthedocs.io/en/latest/modules/loader.html#torch_geometric.loader.NeighborSampler) class. \n","Neighbor sampling is based on building a node’s computation graph without storing irrelevant information for that given node, thus, making it more efficient. Each node produces a single computation graph, where for each node in a k-hop neighborhood, at most, $H_k$ neighbors are randomly sampled. Each node's  computation graph will therefore involve $\\prod^K_{k=1} H_k$ leaf nodes for a K-layer GNN. \n","\n","The successive layers of each node's computation graph can be conceptualized as bi-partite graphs, where each bi-partite graph represents the information flow in one layer of message passing (shown in figure below). Let us look at how one message passing layer is represented through a bi-partite graph. For layer 3, the blue (or black) dots are the source nodes needed to compute message passing (in layer 3 of our GNN) to produce updated embeddings for the target nodes (shown in red). In the bi-partite represention, information flows from left to right, where we highlight that the left hand side of the graph specifically includes the target nodes to allow for skip-connections and added self-loops. Additionally, we stress that moving from one layer to the next, the bi-partite graphs are explicitly constructed so the target nodes for layer $k-1$ are the needed source nodes for layer $k$, where the output of the final layer is exactly the embeddings of the nodes in our minibatch.\n","\n","Note, when constructing mini-batches, individual node computation graphs are combined to create this bi-partite structure. If you'd like to learn more about information on neighborhood sampling, this\n","**[blog](https://towardsdatascience.com/sampling-large-graphs-in-pytorch-geometric-97a6119c41f9)** provides an excellent description.\n","\n","![img]( https://drive.google.com/uc?export=view&id=1QqcrEsN-HpSHgwHiOD4Dh6yIawZh0Pgj)\n","\n","\n","**PyG Docs**\n","\n","The neighbor sampler from the “Inductive Representation Learning on Large Graphs” paper, which allows for mini-batch training of GNNs on large-scale graphs where full-batch training is not feasible.\n","\n","Given a GNN with  layers and a specific mini-batch of nodes `node_idx` **for which we want to compute embedding**s, this module **iteratively samples** neighbors (from right to left) and **constructs bipartite graphs**(from left to right) that simulate the actual computation flow of GNNs.\n","\n","More specifically, **sizes** denotes how much neighbors we want to sample for each node in each layer. This module then takes in these sizes and iteratively samples sizes[l] for each node involved in layer `l`. In the next layer, sampling is repeated for the union of nodes that were already encountered. The actual computation graphs are then returned **in reverse-mode**, meaning that we pass messages from a larger set of nodes to a smaller one, until we reach the nodes for which we originally wanted to compute embeddings.\n","\n","Hence, an item returned by NeighborSampler holds the current batch_size, the IDs n_id of all nodes involved in the computation, and a list of bipartite graph objects via the tuple (edge_index, e_id, size), where edge_index represents the bipartite edges between source and target nodes, e_id denotes the IDs of original edges in the full graph, and size holds the shape of the bipartite graph. For each bipartite graph, target nodes are also included at the beginning of the list of source nodes so that one can easily apply skip-connections or add self-loops.\n","\n","\n","If you are interested in memory-efficient aggregations, please refer to PyG's [Memory-Efficient Aggregations](https://pytorch-geometric.readthedocs.io/en/latest/notes/sparse_tensor.html).  Following is an **example** that uses the Neighbor Sampling method on training the OGB `arxiv` dataset. "]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":1107,"status":"ok","timestamp":1657143136955,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"nWlyStlRO6_u","outputId":"573ba38e-7508-4ed7-92d3-30833e9963ae"},"outputs":[{"output_type":"stream","name":"stdout","text":["Device: cuda\n"]}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  dataset_name = 'ogbn-arxiv'\n","  dataset = PygNodePropPredDataset(name=dataset_name,\n","                                  transform=T.ToSparseTensor())\n","  data = dataset[0]\n","  data.adj_t = data.adj_t.to_symmetric()\n","  device = 'cuda' if torch.cuda.is_available() else 'cpu'\n","\n","  print('Device: {}'.format(device))\n","\n","  data = data.to(device)\n","  split_idx = dataset.get_idx_split()\n","  train_idx = split_idx['train'].to(device)\n","\n","  sampled_subgraph_batch_loader = None\n","  full_subgraph_loader = None\n","  \n","  ############# Your code here ############\n","  ## (~2 line of code)\n","  ## Note:\n","  ## 1. Construct the NeighborSampler `sampled_subgraph_batch_loader`. \n","  ##    Use a batch size of 4096, turn shuffle on, and only \n","  ##    use train_idx nodes to create mini-batches. During sampling,\n","  ##    sample up to 10 neighbors in layer one and 5 neighbors in layer 2.\n","  ## 2. Construct the NeighborSampler `full_subgraph_loader`. \n","  ##    Use a batch size of 4096 and turn shuffle off. Sample all neighbors\n","  ##    for only ONE layer and consider all nodes for sampling mini-batches!\n","  ##    We use this loader for the inference / test phase of our model. In the\n","  ##    inference function we will why we only need to sample complete\n","  ##    1-hop neighborhoods.\n","  ## 3. Look at the NeighborSampler documentation to figure out which \n","  ##    parameters you need to set:\n","  ##    https://pytorch-geometric.readthedocs.io/en/latest/modules/loader.html#torch_geometric.loader.NeighborSampler\n","  sampled_subgraph_batch_loader = NeighborSampler(data.adj_t, node_idx=train_idx, \n","                               sizes=[10, 5], batch_size=4096, \n","                               shuffle=True, num_workers=2)\n","  full_subgraph_loader = NeighborSampler(data.adj_t, node_idx=None, \n","                               sizes=[-1], batch_size=4096, \n","                               shuffle=False, num_workers=2)\n","  #################################################################################\n","\n","  evaluator = Evaluator(name='ogbn-arxiv')"]},{"cell_type":"markdown","metadata":{"id":"VjdkIcFpRYyl"},"source":["## GNN Model\n","\n","After creating your `NeighborSampler`, you also need to **modify your model to support the mini-batch training**.\n","\n","The `forward` function will take the node feature `x` and a list of three-element **tuples** `adjs`. The length of `adjs` is the number of layers of GNN. Each element in `adjs` contains following elements:\n","* `edge_index`: The **edge index** tensor between source and destination nodes, which **forms a bipartite graph**.\n","* `e_id`: The indices of the edges in the original graph.\n","* `size`: The **shape of the bipartite graph**, in (*number of source nodes*, *number of destination nodes*) format.\n"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"gRBJS_5qRWbu"},"outputs":[],"source":["class SAGE(torch.nn.Module):\n","    def __init__(self, input_dim, hidden_dim, output_dim, num_layers,\n","                 dropout):\n","        super(SAGE, self).__init__()\n","\n","        self.convs = torch.nn.ModuleList()\n","        self.bns = torch.nn.ModuleList()\n","\n","        self.convs.append(SAGEConv(input_dim, hidden_dim))\n","        self.bns.append(torch.nn.BatchNorm1d(hidden_dim))\n","\n","        for i in range(num_layers - 2):\n","            self.convs.append(\n","                SAGEConv(hidden_dim, hidden_dim))\n","            self.bns.append(torch.nn.BatchNorm1d(hidden_dim))\n","        self.convs.append(SAGEConv(hidden_dim, output_dim))\n","\n","        self.softmax = torch.nn.LogSoftmax(dim=1)\n","\n","        self.dropout = dropout\n","\n","        self.num_layers = num_layers\n","\n","    def reset_parameters(self):\n","        for conv in self.convs:\n","            conv.reset_parameters()\n","        for bn in self.bns:\n","            bn.reset_parameters()\n","\n","    def forward(self, x, adjs, mode=\"batch\"):\n","        if mode == \"batch\":# mini-batch case\n","            ############# Your code here ############\n","            ## (~7 line of code)\n","            ## Note:\n","            ## 1. Loop through the list `adjs` and apply L GNN layers.\n","            ##    Refer to the description above for the elements in each tuple\n","            ##    adjs[l].\n","            ## 2. Our GNN model is of the form:\n","            ##      conv -> bn -> relu -> dropout -> ... -> conv\n","            ## 3. As described above, each layer is defined by a bipartite graph\n","            ##    between (source nodes and target nodes), where the size parameter\n","            ##    tells us how many source nodes and target nodes we have. Thus, \n","            ##    size is a list of length 2.\n","            ## 4. Rather than passing just x to the SAGEConv layer, you can pass\n","            ##    a tuple of the form (x_src, x_target). With this formulation\n","            ##    we only produce embeddings for the `x_target` nodes and use\n","            ##    `x_source` as the nodes needed for message passing.  \n","            ##\n","            ##    Hint: \n","            ##      - Target nodes (red) are included as the first nodes in the source nodes(blue). For self loops or skip connection\n","            ##      - The target nodes for layer (l) become the source nodes \n","            ##      for layer (l+1)!\n","            for i, (edge_index, _, size) in enumerate(adjs): # loop over each layer\n","              x_target = x[:size[1]]  # the main difference: get target node features from source nodes. Target nodes are always placed first.\n","              x = self.convs[i]((x, x_target), edge_index) # the main difference: conv layer\n","              if i != self.num_layers - 1: # conv -> bn -> relu -> dropout\n","                  x = self.bns[i](x)\n","                  x = F.relu(x)\n","                  x = F.dropout(x, p=self.dropout, training=self.training)\n","            #####################################\n","        else:# full-batch case\n","            for i, conv in enumerate(self.convs): # loop over each layer\n","                x = conv(x, adjs)\n","                if i != self.num_layers - 1:\n","                    x = self.bns[i](x)\n","                    x = F.relu(x)\n","                    x = F.dropout(x, p=self.dropout, training=self.training)\n","        return self.softmax(x)\n","    \n","    def inference(self, x_all, all_loader):\n","        # This function will be called in test\n","        for i in range(self.num_layers):\n","            xs = []\n","            ############# Your code here ############\n","            ## (~7 line of code)\n","            ## Note:\n","            ## 1. Very similar idea to the forward function!\n","            ## 2. Looping through all_loader to apply conv layer i to each\n","            ##    batch of nodes, where all_loader returns tuples of:\n","            ##      - batch_size\n","            ##      - node_indeces of the nodes we want to sample from x_all\n","            ##      in the current batch.\n","            ##      - adj: see above for what is in the tuple adj.\n","            ## 3. After selecting nodes using the node_indeces, follow the same\n","            ##    steps as the forward function.\n","            ## 4. Since we are doing mini-batches of nodes, we now need\n","            ##    to append the generated embeddings to compute all\n","            ##    the node embeddings!\n","            ## 5. Rember to move `adj` and the selected nodes for each batch to\n","            ##    the GPU `device`.\n","            for batch_size, node_indeces, adj in all_loader:\n","              edge_index, _, size = adj.to(device)\n","              x = x_all[node_indeces.to(device)].to(device) # select sample nodes\n","              # follow the forward function\n","              x_target = x[:size[1]]  # get target node features from source nodes. Target nodes are always placed first.\n","              x = self.convs[i]((x, x_target), edge_index.to(device))\n","              if i != self.num_layers - 1:\n","                  x = self.bns[i](x)\n","                  x = F.relu(x)\n","                  x = F.dropout(x, p=self.dropout, training=self.training)\n","              xs.append(x[:batch_size].cpu()) # only take the first target node embeddings(red)\n","            #####################################\n","            \n","            # Concatenate all updated embeddings xs into one tensor.\n","            # We simulate the update process at the end of message\n","            # passing. Because of this we only have to sample 1-hop neighborhoods\n","            # for our full_subgraph_loader!\n","            x_all = torch.cat(xs, dim=0)\n","\n","        return self.softmax(x_all)"]},{"cell_type":"markdown","metadata":{"id":"7cfm7K3wRqqY"},"source":["## Training and Testing\n","\n","Now we provide the training and testing functions for you.\n","\n","In both training and testing, we need to sample batches from the dataloader.\n","\n","Each batch in the `NeighborSampler` dataloader holds three elements:\n","* `batch_size`: The batch size specified in the dataloader.\n","* `n_id`: All nodes (in index format) used in the adjacency matrices.\n","* `adjs`: The three-element tuples."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"-JN0-_QCRn8N"},"outputs":[],"source":["def train(model, data, train_loader, train_idx, optimizer, loss_fn, mode=\"batch\"):\n","    model.train()\n","\n","    total_loss = 0\n","    if mode == \"batch\":\n","        for batch_size, n_id, adjs in train_loader:\n","            # Move all adj sparse tensors to GPU\n","            adjs = [adj.to(device) for adj in adjs]\n","            optimizer.zero_grad()\n","\n","            # Index on the node features\n","            out = model(data.x[n_id], adjs)\n","            train_label = data.y[n_id[:batch_size]].squeeze(-1)\n","            loss = loss_fn(out, train_label)\n","            loss.backward()\n","            optimizer.step()\n","            total_loss += loss.item()\n","    else:\n","        optimizer.zero_grad()\n","        out = model(data.x, data.adj_t, mode=mode)[train_idx]\n","        train_label = data.y.squeeze(1)[train_idx]\n","        loss = loss_fn(out, train_label)\n","        loss.backward()\n","        optimizer.step()\n","        total_loss = loss.item()\n","\n","    return total_loss\n","\n","@torch.no_grad()\n","def test(model, data, all_loader, split_idx, evaluator, mode=\"batch\", save_model_results=False):\n","    model.eval()\n","\n","    if mode == \"batch\":\n","        out = model.inference(data.x, all_loader)\n","    else:\n","        out = model(data.x, data.adj_t, mode=\"all\")\n","\n","    y_true = data.y.cpu()\n","    y_pred = out.argmax(dim=-1, keepdim=True)\n","\n","    train_acc = evaluator.eval({\n","        'y_true': y_true[split_idx['train']],\n","        'y_pred': y_pred[split_idx['train']],\n","    })['acc']\n","    valid_acc = evaluator.eval({\n","        'y_true': y_true[split_idx['valid']],\n","        'y_pred': y_pred[split_idx['valid']],\n","    })['acc']\n","    test_acc = evaluator.eval({\n","        'y_true': y_true[split_idx['test']],\n","        'y_pred': y_pred[split_idx['test']],\n","    })['acc']\n","\n","    if save_model_results:\n","      print (\"Saving Model Predictions\")\n","\n","      data = {}\n","      data['y_pred'] = y_pred.view(-1).cpu().detach().numpy()\n","\n","      df = pd.DataFrame(data=data)\n","      # Save locally as csv\n","      df.to_csv('ogbn-arxiv_' + mode + '.csv', sep=',', index=False)\n","\n","    return train_acc, valid_acc, test_acc"]},{"cell_type":"markdown","metadata":{"id":"AiehZ8OiR2q9"},"source":["## Mini-batch Training\n","\n","Test your model using mini-batch training, based on our NeighborSampler!"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"zFaI2eCARy0v","outputId":"9838c89f-f457-410f-83d3-cfa31a03faeb","executionInfo":{"status":"ok","timestamp":1657143690321,"user_tz":-120,"elapsed":553022,"user":{"displayName":"Min Wang","userId":"04005434052037859847"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch: 01, Loss: 40.0729, Train: 63.44%, Valid: 64.08% Test: 63.04%\n","Epoch: 02, Loss: 28.7527, Train: 67.32%, Valid: 66.62% Test: 64.96%\n","Epoch: 03, Loss: 26.6194, Train: 68.69%, Valid: 68.53% Test: 68.23%\n","Epoch: 04, Loss: 25.5883, Train: 69.78%, Valid: 69.26% Test: 68.75%\n","Epoch: 05, Loss: 25.0509, Train: 70.72%, Valid: 69.37% Test: 68.26%\n","Epoch: 06, Loss: 24.6593, Train: 70.58%, Valid: 69.05% Test: 67.53%\n","Epoch: 07, Loss: 24.2584, Train: 71.28%, Valid: 69.65% Test: 68.87%\n","Epoch: 08, Loss: 23.9959, Train: 71.37%, Valid: 69.50% Test: 68.45%\n","Epoch: 09, Loss: 23.8515, Train: 72.07%, Valid: 69.35% Test: 67.86%\n","Epoch: 10, Loss: 23.5434, Train: 72.26%, Valid: 70.23% Test: 69.35%\n","Epoch: 11, Loss: 23.3411, Train: 72.56%, Valid: 70.40% Test: 69.68%\n","Epoch: 12, Loss: 23.1937, Train: 72.65%, Valid: 69.41% Test: 67.92%\n","Epoch: 13, Loss: 23.0489, Train: 72.93%, Valid: 69.13% Test: 67.15%\n","Epoch: 14, Loss: 22.9534, Train: 73.18%, Valid: 69.98% Test: 67.91%\n","Epoch: 15, Loss: 22.8220, Train: 73.56%, Valid: 70.47% Test: 69.44%\n","Epoch: 16, Loss: 22.7278, Train: 73.46%, Valid: 70.56% Test: 69.57%\n","Epoch: 17, Loss: 22.4424, Train: 73.84%, Valid: 70.46% Test: 69.27%\n","Epoch: 18, Loss: 22.5012, Train: 73.70%, Valid: 69.59% Test: 67.79%\n","Epoch: 19, Loss: 22.5065, Train: 73.48%, Valid: 70.39% Test: 69.32%\n","Epoch: 20, Loss: 22.2319, Train: 73.95%, Valid: 69.54% Test: 67.37%\n","Epoch: 21, Loss: 22.3019, Train: 73.66%, Valid: 70.52% Test: 69.98%\n","Epoch: 22, Loss: 22.1812, Train: 74.13%, Valid: 70.57% Test: 69.80%\n","Epoch: 23, Loss: 21.9962, Train: 74.12%, Valid: 70.26% Test: 69.08%\n","Epoch: 24, Loss: 21.9284, Train: 74.34%, Valid: 71.17% Test: 70.13%\n","Epoch: 25, Loss: 21.9113, Train: 74.12%, Valid: 70.30% Test: 69.13%\n","Epoch: 26, Loss: 21.8488, Train: 74.61%, Valid: 70.40% Test: 68.87%\n","Epoch: 27, Loss: 21.7777, Train: 74.47%, Valid: 70.23% Test: 69.18%\n","Epoch: 28, Loss: 21.7567, Train: 74.99%, Valid: 70.96% Test: 69.91%\n","Epoch: 29, Loss: 21.6714, Train: 74.91%, Valid: 70.27% Test: 68.88%\n","Epoch: 30, Loss: 21.5037, Train: 75.10%, Valid: 70.67% Test: 69.46%\n","Epoch: 31, Loss: 21.5987, Train: 75.18%, Valid: 70.86% Test: 69.74%\n","Epoch: 32, Loss: 21.5635, Train: 74.91%, Valid: 69.52% Test: 67.44%\n","Epoch: 33, Loss: 21.5953, Train: 74.98%, Valid: 70.25% Test: 68.71%\n","Epoch: 34, Loss: 21.5855, Train: 75.32%, Valid: 70.42% Test: 69.14%\n","Epoch: 35, Loss: 21.4338, Train: 75.28%, Valid: 71.05% Test: 70.41%\n","Epoch: 36, Loss: 21.3743, Train: 75.04%, Valid: 70.43% Test: 69.71%\n","Epoch: 37, Loss: 21.4493, Train: 75.48%, Valid: 70.70% Test: 69.76%\n","Epoch: 38, Loss: 21.3068, Train: 75.50%, Valid: 70.42% Test: 69.40%\n","Epoch: 39, Loss: 21.3681, Train: 75.52%, Valid: 70.82% Test: 69.61%\n","Epoch: 40, Loss: 21.3647, Train: 75.66%, Valid: 70.54% Test: 69.03%\n","Epoch: 41, Loss: 21.2395, Train: 75.58%, Valid: 70.84% Test: 69.51%\n","Epoch: 42, Loss: 21.2585, Train: 75.65%, Valid: 70.30% Test: 69.03%\n","Epoch: 43, Loss: 21.1365, Train: 75.75%, Valid: 71.02% Test: 69.53%\n","Epoch: 44, Loss: 21.0814, Train: 75.88%, Valid: 70.88% Test: 69.91%\n","Epoch: 45, Loss: 21.0406, Train: 75.73%, Valid: 70.51% Test: 69.10%\n","Epoch: 46, Loss: 21.0869, Train: 75.70%, Valid: 70.47% Test: 69.23%\n","Epoch: 47, Loss: 21.0159, Train: 76.16%, Valid: 70.73% Test: 69.27%\n","Epoch: 48, Loss: 20.8848, Train: 75.98%, Valid: 70.89% Test: 69.69%\n","Epoch: 49, Loss: 21.0971, Train: 76.25%, Valid: 70.74% Test: 69.45%\n","Epoch: 50, Loss: 20.9096, Train: 76.00%, Valid: 70.65% Test: 69.63%\n","Epoch: 51, Loss: 20.9190, Train: 75.92%, Valid: 70.02% Test: 68.01%\n","Epoch: 52, Loss: 20.8559, Train: 75.99%, Valid: 70.67% Test: 69.26%\n","Epoch: 53, Loss: 20.8412, Train: 76.12%, Valid: 70.85% Test: 69.35%\n","Epoch: 54, Loss: 20.9138, Train: 76.06%, Valid: 71.05% Test: 69.77%\n","Epoch: 55, Loss: 20.9040, Train: 76.00%, Valid: 70.23% Test: 68.95%\n","Epoch: 56, Loss: 20.8415, Train: 76.36%, Valid: 70.31% Test: 68.60%\n","Epoch: 57, Loss: 20.9010, Train: 76.29%, Valid: 71.01% Test: 70.06%\n","Epoch: 58, Loss: 20.7705, Train: 76.20%, Valid: 70.74% Test: 68.87%\n","Epoch: 59, Loss: 20.8062, Train: 76.39%, Valid: 71.04% Test: 69.72%\n","Epoch: 60, Loss: 20.7540, Train: 76.37%, Valid: 70.81% Test: 69.76%\n","Epoch: 61, Loss: 20.7259, Train: 76.45%, Valid: 70.73% Test: 69.14%\n","Epoch: 62, Loss: 20.7215, Train: 76.53%, Valid: 70.69% Test: 69.24%\n","Epoch: 63, Loss: 20.5922, Train: 76.35%, Valid: 70.46% Test: 68.80%\n","Epoch: 64, Loss: 20.7113, Train: 76.34%, Valid: 70.86% Test: 69.64%\n","Epoch: 65, Loss: 20.6449, Train: 76.45%, Valid: 70.99% Test: 69.63%\n","Epoch: 66, Loss: 20.4825, Train: 76.77%, Valid: 70.69% Test: 69.73%\n","Epoch: 67, Loss: 20.6953, Train: 76.61%, Valid: 70.89% Test: 69.80%\n","Epoch: 68, Loss: 20.5725, Train: 76.59%, Valid: 70.65% Test: 69.55%\n","Epoch: 69, Loss: 20.6279, Train: 76.49%, Valid: 70.47% Test: 68.73%\n","Epoch: 70, Loss: 20.6282, Train: 76.60%, Valid: 71.08% Test: 69.81%\n","Epoch: 71, Loss: 20.5985, Train: 76.50%, Valid: 70.89% Test: 70.08%\n","Epoch: 72, Loss: 20.5907, Train: 76.60%, Valid: 70.58% Test: 69.34%\n","Epoch: 73, Loss: 20.6366, Train: 76.59%, Valid: 70.64% Test: 69.60%\n","Epoch: 74, Loss: 20.4848, Train: 76.76%, Valid: 70.51% Test: 69.13%\n","Epoch: 75, Loss: 20.5116, Train: 76.69%, Valid: 70.57% Test: 69.05%\n","Epoch: 76, Loss: 20.5423, Train: 76.67%, Valid: 70.63% Test: 69.22%\n","Epoch: 77, Loss: 20.4112, Train: 76.74%, Valid: 70.85% Test: 69.77%\n","Epoch: 78, Loss: 20.4322, Train: 76.87%, Valid: 70.42% Test: 68.88%\n","Epoch: 79, Loss: 20.5054, Train: 76.93%, Valid: 70.83% Test: 69.94%\n","Epoch: 80, Loss: 20.3866, Train: 76.91%, Valid: 70.87% Test: 69.72%\n","Epoch: 81, Loss: 20.3771, Train: 76.97%, Valid: 70.47% Test: 68.85%\n","Epoch: 82, Loss: 20.4064, Train: 77.03%, Valid: 70.62% Test: 69.46%\n","Epoch: 83, Loss: 20.4883, Train: 77.01%, Valid: 70.48% Test: 69.04%\n","Epoch: 84, Loss: 20.4974, Train: 77.00%, Valid: 70.84% Test: 69.74%\n","Epoch: 85, Loss: 20.4497, Train: 77.07%, Valid: 70.40% Test: 68.53%\n","Epoch: 86, Loss: 20.4707, Train: 76.94%, Valid: 70.94% Test: 70.05%\n","Epoch: 87, Loss: 20.3937, Train: 76.83%, Valid: 70.51% Test: 69.03%\n","Epoch: 88, Loss: 20.3811, Train: 77.06%, Valid: 70.80% Test: 69.68%\n","Epoch: 89, Loss: 20.3204, Train: 77.04%, Valid: 70.66% Test: 69.53%\n","Epoch: 90, Loss: 20.3321, Train: 76.98%, Valid: 70.54% Test: 69.46%\n","Epoch: 91, Loss: 20.2467, Train: 77.26%, Valid: 70.64% Test: 69.06%\n","Epoch: 92, Loss: 20.3014, Train: 76.98%, Valid: 71.04% Test: 70.12%\n","Epoch: 93, Loss: 20.3015, Train: 77.20%, Valid: 70.73% Test: 69.28%\n","Epoch: 94, Loss: 20.2676, Train: 77.11%, Valid: 69.89% Test: 67.73%\n","Epoch: 95, Loss: 20.2261, Train: 77.05%, Valid: 70.82% Test: 69.29%\n","Epoch: 96, Loss: 20.3077, Train: 77.33%, Valid: 70.49% Test: 68.90%\n","Epoch: 97, Loss: 20.2304, Train: 77.33%, Valid: 70.99% Test: 70.02%\n","Epoch: 98, Loss: 20.1700, Train: 77.20%, Valid: 70.57% Test: 69.13%\n","Epoch: 99, Loss: 20.2554, Train: 77.32%, Valid: 70.65% Test: 69.18%\n","Epoch: 100, Loss: 20.3614, Train: 77.42%, Valid: 70.73% Test: 69.47%\n","Saving Model Predictions\n","Best model: Train: 74.34%, Valid: 71.17% Test: 70.13%\n"]}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  args = {\n","      'device': device,\n","      'num_layers': 2,\n","      'hidden_dim': 128,\n","      'dropout': 0.5,\n","      'lr': 0.01,\n","      'epochs': 100,\n","  }\n","\n","  set_seed()\n","\n","  batch_model = SAGE(data.num_features, args['hidden_dim'],\n","              dataset.num_classes, args['num_layers'],\n","              args['dropout']).to(device)\n","  batch_model.reset_parameters()\n","\n","  optimizer = torch.optim.Adam(batch_model.parameters(), lr=args['lr'])\n","  loss_fn = F.nll_loss\n","\n","  best_batch_model = None\n","  best_valid_acc = 0\n","\n","  batch_results = []\n","\n","  for epoch in range(1, 1 + args[\"epochs\"]):\n","      loss = train(batch_model, data, sampled_subgraph_batch_loader, train_idx, optimizer, loss_fn, mode=\"batch\")\n","      result = test(batch_model, data, full_subgraph_loader, split_idx, evaluator, mode=\"batch\")\n","      batch_results.append(result)\n","      train_acc, valid_acc, test_acc = result\n","      if valid_acc > best_valid_acc:\n","          best_valid_acc = valid_acc\n","          best_batch_model = copy.deepcopy(batch_model)\n","      print(f'Epoch: {epoch:02d}, '\n","            f'Loss: {loss:.4f}, '\n","            f'Train: {100 * train_acc:.2f}%, '\n","            f'Valid: {100 * valid_acc:.2f}% '\n","            f'Test: {100 * test_acc:.2f}%')\n","  best_result = test(best_batch_model, data, full_subgraph_loader, split_idx, evaluator, mode=\"batch\", save_model_results=True)\n","  train_acc, valid_acc, test_acc = best_result\n","  print(f'Best model: '\n","        f'Train: {100 * train_acc:.2f}%, '\n","        f'Valid: {100 * valid_acc:.2f}% '\n","        f'Test: {100 * test_acc:.2f}%')"]},{"cell_type":"markdown","metadata":{"id":"JwcRKcAh16RV"},"source":["## **Question 3:** What is the maximum accuracy obtained on the test set using mini-batch training? (15 points)\n","\n","Running the cell above will show the results of your best model and save your best model's predictions to a file named ogbn-arxiv_batch.csv'.\n","\n","As you have seen before you can view this file by clicking on the Folder icon on the left side pannel. When you sumbit your assignment, you will have to download this file and attatch it to your submission."]},{"cell_type":"markdown","metadata":{"id":"-OyqW-1pSMLW"},"source":["## Full-batch Training\n","\n","Now for reference, compare training over all the nodes using full-batch mode (i.e. as we have done in the previous Colabs)."]},{"cell_type":"code","execution_count":null,"metadata":{"executionInfo":{"elapsed":7962,"status":"ok","timestamp":1657143698277,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"mU5eAviTSFMO","colab":{"base_uri":"https://localhost:8080/"},"outputId":"3a0521c7-94a9-41d6-92ad-2aad30546904"},"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch: 01, Loss: 4.0161, Train: 20.83%, Valid: 16.98% Test: 14.02%\n","Epoch: 02, Loss: 2.5924, Train: 35.77%, Valid: 40.95% Test: 43.24%\n","Epoch: 03, Loss: 2.2350, Train: 36.83%, Valid: 42.64% Test: 44.71%\n","Epoch: 04, Loss: 2.0620, Train: 38.98%, Valid: 45.24% Test: 47.67%\n","Epoch: 05, Loss: 1.9249, Train: 41.61%, Valid: 48.75% Test: 51.48%\n","Epoch: 06, Loss: 1.8114, Train: 44.48%, Valid: 51.62% Test: 53.64%\n","Epoch: 07, Loss: 1.7274, Train: 47.47%, Valid: 53.59% Test: 54.83%\n","Epoch: 08, Loss: 1.6563, Train: 50.09%, Valid: 55.24% Test: 55.95%\n","Epoch: 09, Loss: 1.6004, Train: 52.29%, Valid: 56.86% Test: 57.23%\n","Epoch: 10, Loss: 1.5477, Train: 53.87%, Valid: 58.08% Test: 58.04%\n","Epoch: 11, Loss: 1.5007, Train: 54.81%, Valid: 58.61% Test: 58.12%\n","Epoch: 12, Loss: 1.4635, Train: 55.43%, Valid: 58.51% Test: 57.94%\n","Epoch: 13, Loss: 1.4348, Train: 56.12%, Valid: 58.76% Test: 57.72%\n","Epoch: 14, Loss: 1.4091, Train: 56.90%, Valid: 59.08% Test: 57.72%\n","Epoch: 15, Loss: 1.3877, Train: 57.84%, Valid: 59.57% Test: 57.99%\n","Epoch: 16, Loss: 1.3698, Train: 58.85%, Valid: 60.33% Test: 58.68%\n","Epoch: 17, Loss: 1.3531, Train: 59.90%, Valid: 61.03% Test: 59.44%\n","Epoch: 18, Loss: 1.3368, Train: 60.87%, Valid: 61.85% Test: 60.28%\n","Epoch: 19, Loss: 1.3145, Train: 61.69%, Valid: 62.70% Test: 61.30%\n","Epoch: 20, Loss: 1.3040, Train: 62.37%, Valid: 63.38% Test: 62.13%\n","Epoch: 21, Loss: 1.2918, Train: 62.92%, Valid: 63.84% Test: 62.84%\n","Epoch: 22, Loss: 1.2749, Train: 63.39%, Valid: 64.33% Test: 63.31%\n","Epoch: 23, Loss: 1.2675, Train: 63.80%, Valid: 64.59% Test: 63.60%\n","Epoch: 24, Loss: 1.2565, Train: 64.19%, Valid: 64.72% Test: 63.64%\n","Epoch: 25, Loss: 1.2462, Train: 64.51%, Valid: 64.92% Test: 63.58%\n","Epoch: 26, Loss: 1.2378, Train: 64.70%, Valid: 64.95% Test: 63.52%\n","Epoch: 27, Loss: 1.2276, Train: 64.89%, Valid: 65.14% Test: 63.62%\n","Epoch: 28, Loss: 1.2193, Train: 65.10%, Valid: 65.43% Test: 63.86%\n","Epoch: 29, Loss: 1.2108, Train: 65.37%, Valid: 65.77% Test: 64.33%\n","Epoch: 30, Loss: 1.2021, Train: 65.69%, Valid: 66.16% Test: 64.92%\n","Epoch: 31, Loss: 1.1942, Train: 66.08%, Valid: 66.62% Test: 65.51%\n","Epoch: 32, Loss: 1.1911, Train: 66.45%, Valid: 66.89% Test: 66.02%\n","Epoch: 33, Loss: 1.1843, Train: 66.71%, Valid: 67.15% Test: 66.39%\n","Epoch: 34, Loss: 1.1750, Train: 66.92%, Valid: 67.39% Test: 66.64%\n","Epoch: 35, Loss: 1.1727, Train: 67.11%, Valid: 67.49% Test: 66.85%\n","Epoch: 36, Loss: 1.1663, Train: 67.30%, Valid: 67.58% Test: 67.04%\n","Epoch: 37, Loss: 1.1577, Train: 67.38%, Valid: 67.63% Test: 67.11%\n","Epoch: 38, Loss: 1.1536, Train: 67.48%, Valid: 67.74% Test: 67.19%\n","Epoch: 39, Loss: 1.1458, Train: 67.57%, Valid: 67.84% Test: 67.17%\n","Epoch: 40, Loss: 1.1418, Train: 67.72%, Valid: 67.93% Test: 67.16%\n","Epoch: 41, Loss: 1.1382, Train: 67.90%, Valid: 68.01% Test: 67.18%\n","Epoch: 42, Loss: 1.1343, Train: 68.07%, Valid: 68.14% Test: 67.31%\n","Epoch: 43, Loss: 1.1295, Train: 68.22%, Valid: 68.27% Test: 67.55%\n","Epoch: 44, Loss: 1.1215, Train: 68.35%, Valid: 68.43% Test: 67.78%\n","Epoch: 45, Loss: 1.1199, Train: 68.48%, Valid: 68.57% Test: 67.92%\n","Epoch: 46, Loss: 1.1174, Train: 68.64%, Valid: 68.69% Test: 67.95%\n","Epoch: 47, Loss: 1.1095, Train: 68.74%, Valid: 68.74% Test: 67.90%\n","Epoch: 48, Loss: 1.1087, Train: 68.88%, Valid: 68.75% Test: 67.90%\n","Epoch: 49, Loss: 1.1059, Train: 69.01%, Valid: 68.78% Test: 67.89%\n","Epoch: 50, Loss: 1.1025, Train: 69.11%, Valid: 68.89% Test: 67.83%\n","Epoch: 51, Loss: 1.0986, Train: 69.23%, Valid: 68.94% Test: 67.95%\n","Epoch: 52, Loss: 1.0921, Train: 69.28%, Valid: 69.00% Test: 68.16%\n","Epoch: 53, Loss: 1.0874, Train: 69.36%, Valid: 69.06% Test: 68.32%\n","Epoch: 54, Loss: 1.0855, Train: 69.48%, Valid: 69.13% Test: 68.29%\n","Epoch: 55, Loss: 1.0844, Train: 69.57%, Valid: 69.09% Test: 68.39%\n","Epoch: 56, Loss: 1.0835, Train: 69.70%, Valid: 69.16% Test: 68.42%\n","Epoch: 57, Loss: 1.0835, Train: 69.73%, Valid: 69.13% Test: 68.47%\n","Epoch: 58, Loss: 1.0752, Train: 69.85%, Valid: 69.08% Test: 68.48%\n","Epoch: 59, Loss: 1.0746, Train: 69.92%, Valid: 69.16% Test: 68.44%\n","Epoch: 60, Loss: 1.0718, Train: 70.01%, Valid: 69.20% Test: 68.49%\n","Epoch: 61, Loss: 1.0686, Train: 70.05%, Valid: 69.21% Test: 68.57%\n","Epoch: 62, Loss: 1.0659, Train: 70.10%, Valid: 69.27% Test: 68.69%\n","Epoch: 63, Loss: 1.0635, Train: 70.18%, Valid: 69.28% Test: 68.71%\n","Epoch: 64, Loss: 1.0618, Train: 70.26%, Valid: 69.31% Test: 68.62%\n","Epoch: 65, Loss: 1.0579, Train: 70.34%, Valid: 69.27% Test: 68.64%\n","Epoch: 66, Loss: 1.0552, Train: 70.44%, Valid: 69.31% Test: 68.62%\n","Epoch: 67, Loss: 1.0539, Train: 70.53%, Valid: 69.41% Test: 68.76%\n","Epoch: 68, Loss: 1.0502, Train: 70.57%, Valid: 69.51% Test: 68.89%\n","Epoch: 69, Loss: 1.0505, Train: 70.64%, Valid: 69.57% Test: 68.98%\n","Epoch: 70, Loss: 1.0480, Train: 70.69%, Valid: 69.65% Test: 69.11%\n","Epoch: 71, Loss: 1.0459, Train: 70.70%, Valid: 69.69% Test: 69.13%\n","Epoch: 72, Loss: 1.0457, Train: 70.75%, Valid: 69.68% Test: 69.13%\n","Epoch: 73, Loss: 1.0415, Train: 70.79%, Valid: 69.68% Test: 69.17%\n","Epoch: 74, Loss: 1.0378, Train: 70.82%, Valid: 69.78% Test: 69.20%\n","Epoch: 75, Loss: 1.0389, Train: 70.83%, Valid: 69.82% Test: 69.27%\n","Epoch: 76, Loss: 1.0325, Train: 70.90%, Valid: 69.76% Test: 69.25%\n","Epoch: 77, Loss: 1.0357, Train: 70.97%, Valid: 69.83% Test: 69.27%\n","Epoch: 78, Loss: 1.0268, Train: 71.08%, Valid: 69.88% Test: 69.15%\n","Epoch: 79, Loss: 1.0246, Train: 71.12%, Valid: 69.87% Test: 69.11%\n","Epoch: 80, Loss: 1.0283, Train: 71.12%, Valid: 69.84% Test: 69.13%\n","Epoch: 81, Loss: 1.0240, Train: 71.12%, Valid: 69.86% Test: 69.20%\n","Epoch: 82, Loss: 1.0223, Train: 71.17%, Valid: 69.86% Test: 69.22%\n","Epoch: 83, Loss: 1.0212, Train: 71.21%, Valid: 69.88% Test: 69.23%\n","Epoch: 84, Loss: 1.0171, Train: 71.24%, Valid: 69.91% Test: 69.17%\n","Epoch: 85, Loss: 1.0203, Train: 71.29%, Valid: 69.91% Test: 69.16%\n","Epoch: 86, Loss: 1.0132, Train: 71.30%, Valid: 70.01% Test: 69.20%\n","Epoch: 87, Loss: 1.0117, Train: 71.33%, Valid: 70.00% Test: 69.23%\n","Epoch: 88, Loss: 1.0085, Train: 71.38%, Valid: 69.99% Test: 69.22%\n","Epoch: 89, Loss: 1.0068, Train: 71.46%, Valid: 70.09% Test: 69.27%\n","Epoch: 90, Loss: 1.0067, Train: 71.57%, Valid: 70.17% Test: 69.24%\n","Epoch: 91, Loss: 1.0111, Train: 71.58%, Valid: 70.16% Test: 69.22%\n","Epoch: 92, Loss: 1.0051, Train: 71.63%, Valid: 70.12% Test: 69.24%\n","Epoch: 93, Loss: 0.9998, Train: 71.65%, Valid: 70.08% Test: 69.21%\n","Epoch: 94, Loss: 1.0013, Train: 71.67%, Valid: 70.15% Test: 69.32%\n","Epoch: 95, Loss: 1.0014, Train: 71.69%, Valid: 70.11% Test: 69.44%\n","Epoch: 96, Loss: 0.9999, Train: 71.71%, Valid: 70.18% Test: 69.57%\n","Epoch: 97, Loss: 0.9972, Train: 71.74%, Valid: 70.14% Test: 69.63%\n","Epoch: 98, Loss: 0.9943, Train: 71.79%, Valid: 70.15% Test: 69.58%\n","Epoch: 99, Loss: 0.9923, Train: 71.88%, Valid: 70.18% Test: 69.46%\n","Epoch: 100, Loss: 0.9918, Train: 71.98%, Valid: 70.18% Test: 69.47%\n","Best model: Train: 71.71%, Valid: 70.18% Test: 69.57%\n"]}],"source":["if 'IS_GRADESCOPE_ENV' not in os.environ:  \n","  # Use the same parameters for a full-batch training\n","  args = {\n","      'device': device,\n","      'num_layers': 2,\n","      'hidden_dim': 128,\n","      'dropout': 0.5,\n","      'lr': 0.01,\n","      'epochs': 100,\n","  }\n","\n","  set_seed()\n","\n","  all_model = SAGE(data.num_features, args['hidden_dim'],\n","              dataset.num_classes, args['num_layers'],\n","              args['dropout']).to(device)\n","  all_model.reset_parameters()\n","\n","  optimizer = torch.optim.Adam(all_model.parameters(), lr=args['lr'])\n","  loss_fn = F.nll_loss\n","\n","  best_all_model = None\n","  best_valid_acc = 0\n","\n","  all_results = []\n","\n","  for epoch in range(1, 1 + args[\"epochs\"]):\n","      # NOTE: For the full batch model, the NeighborSampler loader is not used!\n","      loss = train(all_model, data, sampled_subgraph_batch_loader, train_idx, optimizer, loss_fn, mode=\"all\")\n","      result = test(all_model, data, full_subgraph_loader, split_idx, evaluator, mode=\"all\")\n","      all_results.append(result)\n","      train_acc, valid_acc, test_acc = result\n","      if valid_acc > best_valid_acc:\n","          best_valid_acc = valid_acc\n","          best_all_model = copy.deepcopy(all_model)\n","      print(f'Epoch: {epoch:02d}, '\n","            f'Loss: {loss:.4f}, '\n","            f'Train: {100 * train_acc:.2f}%, '\n","            f'Valid: {100 * valid_acc:.2f}% '\n","            f'Test: {100 * test_acc:.2f}%')\n","  best_result = test(best_all_model, data, full_subgraph_loader, split_idx, evaluator, mode=\"all\")\n","  train_acc, valid_acc, test_acc = best_result\n","  print(f'Best model: '\n","        f'Train: {100 * train_acc:.2f}%, '\n","        f'Valid: {100 * valid_acc:.2f}% '\n","        f'Test: {100 * test_acc:.2f}%')"]},{"cell_type":"markdown","metadata":{"id":"NrECcOQQSZo1"},"source":["## Visualization"]},{"cell_type":"code","execution_count":null,"metadata":{"executionInfo":{"elapsed":598,"status":"ok","timestamp":1657143698870,"user":{"displayName":"Min Wang","userId":"04005434052037859847"},"user_tz":-120},"id":"sh_qvSG1SV63","colab":{"base_uri":"https://localhost:8080/","height":458},"outputId":"a935ec29-9500-4788-f201-26a38aa917c7"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 648x504 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}],"source":["import numpy as np\n","from matplotlib import pyplot as plt\n","\n","if 'IS_GRADESCOPE_ENV' not in os.environ:\n","  batch_results = np.array(batch_results)\n","  all_results = np.array(all_results)\n","\n","  x = np.arange(1, 101)\n","\n","  plt.figure(figsize=(9, 7))\n","\n","  plt.plot(x, batch_results[:, 1], label=\"Batch Validation\")\n","  plt.plot(x, batch_results[:, 2], label=\"Batch Test\")\n","  plt.plot(x, all_results[:, 1], label=\"All Validation\")\n","  plt.plot(x, all_results[:, 2], label=\"All Test\")\n","  plt.title('Model Accuracy')\n","  plt.ylabel('Accuracy')\n","  plt.xlabel('Epoch')\n","  plt.legend()\n","  plt.show()"]},{"cell_type":"markdown","metadata":{"id":"e7JXsMTBgeOI"},"source":["# Submission\n","\n","You will need to submit three files on Gradescope to complete this notebook. \n","\n","1.   Your completed *XCS224W_Colab4.ipynb*. From the \"File\" menu select \"Download .ipynb\" to save a local copy of your completed Colab. \n","2.  *ACM-Node-MeanAgg.csv* \n","3.  *ACM-Node-AttentionAgg.csv*\n","4.  *ogbn-arxiv_batch.csv* \n","\n","Download the csv files by selecting the *Folder* icon on the left panel. \n","\n","To submit your work, zip the files downloaded in steps 1-3 above and submit to gradescope. **NOTE:** DO NOT rename any of the downloaded files. "]},{"cell_type":"code","execution_count":null,"metadata":{"id":"Y_vcsDcYVaGI"},"outputs":[],"source":[""]}],"metadata":{"accelerator":"GPU","colab":{"collapsed_sections":[],"machine_shape":"hm","name":"XCS224W_Colab4.ipynb","provenance":[{"file_id":"1TEVP2989Km8pZ-YrQu0ec2_v38uKJX23","timestamp":1656233236695}]},"gpuClass":"standard","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":"3.8.3"}},"nbformat":4,"nbformat_minor":0}