jjfraaa
/
AutoGL
Not watched
1
0
Fork 0
Code
Releases 3 Wiki evaluate Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain HPC
Browse Source

notebook for link prediction

develop/0.4/predevelop
蔡婕 3 years ago
parent
eca9d0567a
commit
b4ef552628
2 changed files with 666 additions and 0 deletions
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. +628
    -0
      examples/ipynb/graph_nas.ipynb
  2. +38
    -0
      examples/ipynb/link_prediction.ipynb

+ 628
- 0
examples/ipynb/graph_nas.ipynb View File

@@ -0,0 +1,628 @@
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Link Prediction\n",
"In this file we will give you a simple example to show how to use AutoGL to do link prediction.\n",
"\n",
"## Import libraries\n",
"First, you should import some libraries and you can set the random seed before you train the model."
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"from autogl.datasets import build_dataset_from_name\n",
"from autogl.solver import AutoLinkPredictor\n",
"from autogl.solver.utils import set_seed\n",
"import argparse\n",
"from autogl.backend import DependentBackend\n",
"from autogl.datasets.utils import split_edges\n",
"from autogl.module.train.evaluation import Auc\n",
"import yaml\n",
"import random\n",
"import torch\n",
"import numpy as np\n",
"from argparse import ArgumentParser, ArgumentDefaultsHelpFormatter\n",
"set_seed(202106)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Config & Dataset\n",
"Then, you can load dataset and graph nas methods from configs"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" NumNodes: 2708\n",
" NumEdges: 10556\n",
" NumFeats: 1433\n",
" NumClasses: 7\n",
" NumTrainingSamples: 140\n",
" NumValidationSamples: 500\n",
" NumTestSamples: 1000\n",
"Done loading data from cached files.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:02:49] WARNING (root/MainThread) Ignore passed dec since enc is a whole model\n"
]
}
],
"source": [
"def parse(args=None):\n",
" parser = ArgumentParser(\"auto link prediction\", formatter_class=ArgumentDefaultsHelpFormatter)\n",
" parser.add_argument('--config', type=str, default=\"../../configs/lp_gcn_benchmark.yml\")\n",
" parser.add_argument('--dataset', choices=['cora', 'citeseer', 'pubmed'], default='cora', type=str)\n",
" parser.add_argument(\"--hpo\", type=str, default=\"tpe\", help=\"hpo methods\")\n",
" parser.add_argument(\"--max_eval\", type=int, default=50, help=\"max hpo evaluation times\")\n",
" parser.add_argument(\"--seed\", type=int, default=0, help=\"random seed\")\n",
" parser.add_argument(\"--device\", default=0, type=int, help=\"GPU device\")\n",
" args = parser.parse_args(args)\n",
" return args\n",
"\n",
"args = parse('')\n",
"\n",
"dataset = build_dataset_from_name(args.dataset)\n",
"dataset = split_edges(dataset, 0.8, 0.05) # split the edges for dataset\n",
"if DependentBackend.is_dgl(): # add self-loop\n",
" import dgl\n",
" # add self loop to 0\n",
" data = list(dataset[0])\n",
" data[0] = dgl.add_self_loop(data[0])\n",
" dataset = [data]\n",
"configs = yaml.load(open(args.config, \"r\").read(), Loader=yaml.FullLoader)\n",
"configs[\"hpo\"][\"name\"] = args.hpo\n",
"configs[\"hpo\"][\"max_evals\"] = args.max_eval\n",
"solver = AutoLinkPredictor.from_config(configs)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Run "
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:LinkPredictor:Use the default train/val/test ratio in given dataset\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:03] INFO (LinkPredictor/MainThread) Use the default train/val/test ratio in given dataset\n",
"HPO Search Phase:\n",
"\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
" 0%| | 0/50 [00:00<?, ?it/s]INFO:hyperopt.tpe:tpe_transform took 0.218073 seconds\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:03] INFO (hyperopt.tpe/MainThread) tpe_transform took 0.218073 seconds\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:hyperopt.tpe:TPE using 0 trials\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:03] INFO (hyperopt.tpe/MainThread) TPE using 0 trials\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:03] WARNING (root/MainThread) Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:HPO:Parameter: {'trainer': {'max_epoch': 96, 'early_stopping_round': 38, 'lr': 0.038766175592297024, 'weight_decay': 2.3821416978310546e-08}, 'encoder': {'num_layers': 3, 'hidden': [111, 236], 'dropout': 0.020573287174414756, 'act': 'leaky_relu'}, 'decoder': {}} auc: 0.8238174623463882 higher_better\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:10] INFO (HPO/MainThread) Parameter: {'trainer': {'max_epoch': 96, 'early_stopping_round': 38, 'lr': 0.038766175592297024, 'weight_decay': 2.3821416978310546e-08}, 'encoder': {'num_layers': 3, 'hidden': [111, 236], 'dropout': 0.020573287174414756, 'act': 'leaky_relu'}, 'decoder': {}} auc: 0.8238174623463882 higher_better\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
" 2%|▏ | 1/50 [00:07<05:52, 7.19s/it]INFO:hyperopt.tpe:tpe_transform took 0.006646 seconds\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:10] INFO (hyperopt.tpe/MainThread) tpe_transform took 0.006646 seconds\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:hyperopt.tpe:TPE using 1/1 trials with best loss -0.823817\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:10] INFO (hyperopt.tpe/MainThread) TPE using 1/1 trials with best loss -0.823817\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:10] WARNING (root/MainThread) Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:HPO:Parameter: {'trainer': {'max_epoch': 92, 'early_stopping_round': 34, 'lr': 0.00927691677178042, 'weight_decay': 7.058659985819521e-10}, 'encoder': {'num_layers': 3, 'hidden': [121, 132], 'dropout': 0.034825764974073435, 'act': 'relu'}, 'decoder': {}} auc: 0.8113790781661261 higher_better\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:13] INFO (HPO/MainThread) Parameter: {'trainer': {'max_epoch': 92, 'early_stopping_round': 34, 'lr': 0.00927691677178042, 'weight_decay': 7.058659985819521e-10}, 'encoder': {'num_layers': 3, 'hidden': [121, 132], 'dropout': 0.034825764974073435, 'act': 'relu'}, 'decoder': {}} auc: 0.8113790781661261 higher_better\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
" 4%|▍ | 2/50 [00:09<03:34, 4.48s/it]INFO:hyperopt.tpe:tpe_transform took 0.003760 seconds\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:13] INFO (hyperopt.tpe/MainThread) tpe_transform took 0.003760 seconds\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:hyperopt.tpe:TPE using 2/2 trials with best loss -0.823817\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:13] INFO (hyperopt.tpe/MainThread) TPE using 2/2 trials with best loss -0.823817\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:13] WARNING (root/MainThread) Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:HPO:Parameter: {'trainer': {'max_epoch': 142, 'early_stopping_round': 35, 'lr': 0.02091397431735606, 'weight_decay': 8.728358614633772e-09}, 'encoder': {'num_layers': 2, 'hidden': [128], 'dropout': 0.14104896295645744, 'act': 'tanh'}, 'decoder': {}} auc: 0.9538308927047588 higher_better\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:24] INFO (HPO/MainThread) Parameter: {'trainer': {'max_epoch': 142, 'early_stopping_round': 35, 'lr': 0.02091397431735606, 'weight_decay': 8.728358614633772e-09}, 'encoder': {'num_layers': 2, 'hidden': [128], 'dropout': 0.14104896295645744, 'act': 'tanh'}, 'decoder': {}} auc: 0.9538308927047588 higher_better\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
" 6%|▌ | 3/50 [00:21<05:55, 7.57s/it]INFO:hyperopt.tpe:tpe_transform took 0.004439 seconds\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:24] INFO (hyperopt.tpe/MainThread) tpe_transform took 0.004439 seconds\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:hyperopt.tpe:TPE using 3/3 trials with best loss -0.953831\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:24] INFO (hyperopt.tpe/MainThread) TPE using 3/3 trials with best loss -0.953831\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:24] WARNING (root/MainThread) Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:HPO:Parameter: {'trainer': {'max_epoch': 132, 'early_stopping_round': 29, 'lr': 0.017963261319463134, 'weight_decay': 1.039920878028204e-09}, 'encoder': {'num_layers': 2, 'hidden': [71], 'dropout': 0.09177502794266801, 'act': 'relu'}, 'decoder': {}} auc: 0.9515678953224188 higher_better\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:29] INFO (HPO/MainThread) Parameter: {'trainer': {'max_epoch': 132, 'early_stopping_round': 29, 'lr': 0.017963261319463134, 'weight_decay': 1.039920878028204e-09}, 'encoder': {'num_layers': 2, 'hidden': [71], 'dropout': 0.09177502794266801, 'act': 'relu'}, 'decoder': {}} auc: 0.9515678953224188 higher_better\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
" 8%|▊ | 4/50 [00:26<05:04, 6.62s/it]INFO:hyperopt.tpe:tpe_transform took 0.006353 seconds\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:29] INFO (hyperopt.tpe/MainThread) tpe_transform took 0.006353 seconds\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:hyperopt.tpe:TPE using 4/4 trials with best loss -0.953831\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:29] INFO (hyperopt.tpe/MainThread) TPE using 4/4 trials with best loss -0.953831\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:29] WARNING (root/MainThread) Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:HPO:Parameter: {'trainer': {'max_epoch': 78, 'early_stopping_round': 37, 'lr': 0.024625236940264132, 'weight_decay': 5.0233612711599956e-08}, 'encoder': {'num_layers': 3, 'hidden': [71, 226], 'dropout': 0.151834136171988, 'act': 'leaky_relu'}, 'decoder': {}} auc: 0.7246650511829877 higher_better\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:35] INFO (HPO/MainThread) Parameter: {'trainer': {'max_epoch': 78, 'early_stopping_round': 37, 'lr': 0.024625236940264132, 'weight_decay': 5.0233612711599956e-08}, 'encoder': {'num_layers': 3, 'hidden': [71, 226], 'dropout': 0.151834136171988, 'act': 'leaky_relu'}, 'decoder': {}} auc: 0.7246650511829877 higher_better\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
" 10%|█ | 5/50 [00:31<04:43, 6.31s/it]INFO:hyperopt.tpe:tpe_transform took 0.003402 seconds\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:35] INFO (hyperopt.tpe/MainThread) tpe_transform took 0.003402 seconds\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:hyperopt.tpe:TPE using 5/5 trials with best loss -0.953831\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:35] INFO (hyperopt.tpe/MainThread) TPE using 5/5 trials with best loss -0.953831\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:35] WARNING (root/MainThread) Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:HPO:Parameter: {'trainer': {'max_epoch': 135, 'early_stopping_round': 30, 'lr': 0.005181891063109031, 'weight_decay': 5.1614599568940116e-08}, 'encoder': {'num_layers': 2, 'hidden': [139], 'dropout': 0.05673383055413166, 'act': 'elu'}, 'decoder': {}} auc: 0.8402327448699992 higher_better\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:38] INFO (HPO/MainThread) Parameter: {'trainer': {'max_epoch': 135, 'early_stopping_round': 30, 'lr': 0.005181891063109031, 'weight_decay': 5.1614599568940116e-08}, 'encoder': {'num_layers': 2, 'hidden': [139], 'dropout': 0.05673383055413166, 'act': 'elu'}, 'decoder': {}} auc: 0.8402327448699992 higher_better\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
" 12%|█▏ | 6/50 [00:35<03:55, 5.35s/it]INFO:hyperopt.tpe:tpe_transform took 0.004568 seconds\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:38] INFO (hyperopt.tpe/MainThread) tpe_transform took 0.004568 seconds\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:hyperopt.tpe:TPE using 6/6 trials with best loss -0.953831\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:38] INFO (hyperopt.tpe/MainThread) TPE using 6/6 trials with best loss -0.953831\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:38] WARNING (root/MainThread) Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:HPO:Parameter: {'trainer': {'max_epoch': 59, 'early_stopping_round': 28, 'lr': 0.017915948926693917, 'weight_decay': 6.174302220575103e-08}, 'encoder': {'num_layers': 2, 'hidden': [195], 'dropout': 0.031863207309905994, 'act': 'tanh'}, 'decoder': {}} auc: 0.9523528331575026 higher_better\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:42] INFO (HPO/MainThread) Parameter: {'trainer': {'max_epoch': 59, 'early_stopping_round': 28, 'lr': 0.017915948926693917, 'weight_decay': 6.174302220575103e-08}, 'encoder': {'num_layers': 2, 'hidden': [195], 'dropout': 0.031863207309905994, 'act': 'tanh'}, 'decoder': {}} auc: 0.9523528331575026 higher_better\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
" 14%|█▍ | 7/50 [00:39<03:26, 4.80s/it]INFO:hyperopt.tpe:tpe_transform took 0.004849 seconds\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:42] INFO (hyperopt.tpe/MainThread) tpe_transform took 0.004849 seconds\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:hyperopt.tpe:TPE using 7/7 trials with best loss -0.953831\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:42] INFO (hyperopt.tpe/MainThread) TPE using 7/7 trials with best loss -0.953831\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[2022-11-09 23:03:42] WARNING (root/MainThread) Ignore passed dec since enc is a whole model\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
" 14%|█▍ | 7/50 [00:40<04:08, 5.78s/it]\n"
]
},
{
"ename": "KeyboardInterrupt",
"evalue": "",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-21-4938daf3de29>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0msolver\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtime_limit\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m3600\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevaluation_method\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mAuc\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mseed\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mseed\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0msolver\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_leaderboard\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mauc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msolver\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mevaluate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmetric\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"auc\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"test auc: {:.4f}\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mauc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/home/jcai/code/AutoGL/autogl/solver/classifier/link_predictor.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, dataset, time_limit, inplace, train_split, val_split, evaluation_method, seed)\u001b[0m\n\u001b[1;32m 362\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 363\u001b[0m optimized, _ = self.hpo_module.optimize(\n\u001b[0;32m--> 364\u001b[0;31m \u001b[0mtrainer\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_compose_dataset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtime_limit\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtime_for_each_model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 365\u001b[0m )\n\u001b[1;32m 366\u001b[0m \u001b[0;31m# to save memory, all the trainer derived will be mapped to cpu\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/home/jcai/code/AutoGL/autogl/module/hpo/advisorbase.py\u001b[0m in \u001b[0;36moptimize\u001b[0;34m(self, trainer, dataset, time_limit, memory_limit)\u001b[0m\n\u001b[1;32m 165\u001b[0m \u001b[0mnew_parameter_values_json\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnew_trial\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparameter_values\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 166\u001b[0m \u001b[0mdecoded_json\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrial_para\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_decode_para\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnew_parameter_values_json\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 167\u001b[0;31m \u001b[0mcurrent_trainer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mperf\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdecoded_json\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 168\u001b[0m \u001b[0mnew_trial\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparameter_values\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdumps\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrial_para\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 169\u001b[0m \u001b[0mnew_trial\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstatus\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"Completed\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/home/jcai/code/AutoGL/autogl/module/hpo/advisorbase.py\u001b[0m in \u001b[0;36mfn\u001b[0;34m(x)\u001b[0m\n\u001b[1;32m 144\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 145\u001b[0m \u001b[0mcurrent_trainer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mduplicate_from_hyper_parameter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 146\u001b[0;31m \u001b[0mcurrent_trainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 147\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_higher_better\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcurrent_trainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_valid_score\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 148\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_higher_better\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/home/jcai/code/AutoGL/autogl/module/train/link_prediction_full.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, dataset, keep_valid_result)\u001b[0m\n\u001b[1;32m 376\u001b[0m \u001b[0;32melif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpyg_dgl\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'dgl'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 377\u001b[0m \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 378\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_train_only_dgl\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 379\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mkeep_valid_result\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 380\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalid_result\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_predict_only_dgl\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/home/jcai/code/AutoGL/autogl/module/train/link_prediction_full.py\u001b[0m in \u001b[0;36m_train_only_dgl\u001b[0;34m(self, data)\u001b[0m\n\u001b[1;32m 314\u001b[0m )\n\u001b[1;32m 315\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 316\u001b[0;31m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 317\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 318\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlr_scheduler_type\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/home/jcai/virtualenvs/py3_hetgnn/lib/python3.6/site-packages/torch/tensor.py\u001b[0m in \u001b[0;36mbackward\u001b[0;34m(self, gradient, retain_graph, create_graph, inputs)\u001b[0m\n\u001b[1;32m 243\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcreate_graph\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 244\u001b[0m inputs=inputs)\n\u001b[0;32m--> 245\u001b[0;31m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mautograd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgradient\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mretain_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 246\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 247\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mregister_hook\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhook\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/home/jcai/virtualenvs/py3_hetgnn/lib/python3.6/site-packages/torch/autograd/__init__.py\u001b[0m in \u001b[0;36mbackward\u001b[0;34m(tensors, grad_tensors, retain_graph, create_graph, grad_variables, inputs)\u001b[0m\n\u001b[1;32m 145\u001b[0m Variable._execution_engine.run_backward(\n\u001b[1;32m 146\u001b[0m \u001b[0mtensors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgrad_tensors_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mretain_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 147\u001b[0;31m allow_unreachable=True, accumulate_grad=True) # allow_unreachable flag\n\u001b[0m\u001b[1;32m 148\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 149\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mKeyboardInterrupt\u001b[0m: "
]
}
],
"source": [
"solver.fit(dataset, time_limit=3600, evaluation_method=[Auc], seed=args.seed)\n",
"solver.get_leaderboard().show()\n",
"auc = solver.evaluate(metric=\"auc\")\n",
"print(\"test auc: {:.4f}\".format(auc))"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3.6.9 ('py3_hetgnn')",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.6.9"
},
"orig_nbformat": 4,
"vscode": {
"interpreter": {
"hash": "c71a81e10bf2a5359758b67f38009e83c4204a5be77a74a1c3689ce1bb3351ac"
}
}
},
"nbformat": 4,
"nbformat_minor": 2
}

+ 38
- 0
examples/ipynb/link_prediction.ipynb View File

@@ -0,0 +1,38 @@
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Link Prediction\n",
"In this file we will give you a simple example to show how to use AutoGL to do link prediction.\n",
"Import libraries\n",
"First, you should import some libraries and you can set the random seed before you train the model."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3.6.9 ('py3_hetgnn')",
"language": "python",
"name": "python3"
},
"language_info": {
"name": "python",
"version": "3.6.9"
},
"orig_nbformat": 4,
"vscode": {
"interpreter": {
"hash": "c71a81e10bf2a5359758b67f38009e83c4204a5be77a74a1c3689ce1bb3351ac"
}
}
},
"nbformat": 4,
"nbformat_minor": 2
}

Write Preview
Loading…
Cancel
Save