From 599303500543b00d4521019438eee281e8daf96b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E9=83=AD=E8=85=BE?= <10195501402@stu.ecnu.edu.cn> Date: Fri, 27 May 2022 15:03:09 +0800 Subject: [PATCH] Upload files to 'notebooks' --- notebooks/data_mining.ipynb | 1 + notebooks/data_pull.ipynb | 306 ++++++++++++++++++++++++++++++++++++++++++++ notebooks/learning.ipynb | 1 + 3 files changed, 308 insertions(+) create mode 100644 notebooks/data_mining.ipynb create mode 100644 notebooks/data_pull.ipynb create mode 100644 notebooks/learning.ipynb diff --git a/notebooks/data_mining.ipynb b/notebooks/data_mining.ipynb new file mode 100644 index 0000000..1ae201e --- /dev/null +++ b/notebooks/data_mining.ipynb @@ -0,0 +1 @@ +{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"Untitled0.ipynb","provenance":[],"authorship_tag":"ABX9TyM0ddfsei1qq14OInyayGof"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"code","execution_count":1,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"FodWVxwnQsoO","executionInfo":{"status":"ok","timestamp":1651554696872,"user_tz":-480,"elapsed":4864,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"1bfdd089-bcfb-4a03-c212-92e9012af972"},"outputs":[{"output_type":"stream","name":"stdout","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"]}],"source":["from google.colab import drive\n","drive.mount('/content/drive')\n","import os"]},{"cell_type":"code","source":["!pip install igraph"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"egxxstKlUbBI","executionInfo":{"status":"ok","timestamp":1651554700151,"user_tz":-480,"elapsed":3284,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"5df62be8-958e-4791-a28f-130dc242d32a"},"execution_count":2,"outputs":[{"output_type":"stream","name":"stdout","text":["Requirement already satisfied: igraph in /usr/local/lib/python3.7/dist-packages (0.9.10)\n","Requirement already satisfied: texttable>=1.6.2 in /usr/local/lib/python3.7/dist-packages (from igraph) (1.6.4)\n"]}]},{"cell_type":"code","source":["from igraph import *"],"metadata":{"id":"_TvuaXmsU2Nm","executionInfo":{"status":"ok","timestamp":1651554700151,"user_tz":-480,"elapsed":7,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":3,"outputs":[]},{"cell_type":"code","source":["target = 'IssueCommentEvent'\n","path = '/content/drive/My Drive/social_computing/data/'+ target + '.txt'"],"metadata":{"id":"ycxti9acRYLt","executionInfo":{"status":"ok","timestamp":1651555382584,"user_tz":-480,"elapsed":544,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":38,"outputs":[]},{"cell_type":"code","source":["f = open(path, encoding = \"utf-8\")\n","file_data = f.readlines()\n","print(len(file_data))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"wA2YbybDRvXs","executionInfo":{"status":"ok","timestamp":1651555383214,"user_tz":-480,"elapsed":3,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"c2369efd-3f18-448f-a04f-7626d4670b43"},"execution_count":39,"outputs":[{"output_type":"stream","name":"stdout","text":["138808\n"]}]},{"cell_type":"code","source":["node_ids = []\n","for x in range(len(file_data)//2-1):\n"," pair = file_data[x*2+1].strip('\\n').split(' ')[1:]\n"," l = len(pair)\n"," for i in range(l-1):\n"," node_ids.append(int(pair[i]))\n","node_ids = list(set(node_ids))"],"metadata":{"id":"m_4GQ8gbPG5Y","executionInfo":{"status":"ok","timestamp":1651555384802,"user_tz":-480,"elapsed":1,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":40,"outputs":[]},{"cell_type":"code","source":["node_ids.sort()\n","print(node_ids[:100])"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"waEiQx0SPdAz","executionInfo":{"status":"ok","timestamp":1651555385317,"user_tz":-480,"elapsed":2,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"2ede948d-48ae-4f43-d015-fda6061094f9"},"execution_count":41,"outputs":[{"output_type":"stream","name":"stdout","text":["[27, 28, 68, 130, 144, 363, 426, 507, 912, 966, 1032, 1252, 1462, 1734, 1904, 2019, 2376, 2548, 2727, 2747, 3045, 3282, 3287, 3314, 3542, 4016, 4048, 4828, 5266, 5497, 5625, 5954, 6040, 6094, 6133, 6738, 7050, 7497, 7661, 8393, 8514, 8970, 9302, 9525, 9664, 9823, 9980, 10664, 10865, 10910, 11137, 11493, 11572, 11573, 11598, 12494, 13677, 13992, 14507, 14603, 15257, 15293, 15422, 15435, 16267, 16309, 16667, 17409, 17504, 18027, 18294, 18639, 19355, 19438, 20669, 20693, 20724, 20954, 21653, 21829, 21950, 22361, 22601, 22826, 23393, 23513, 23647, 23715, 23892, 23916, 23931, 24954, 25348, 26288, 26554, 27039, 27372, 27412, 28039, 28438]\n"]}]},{"cell_type":"code","source":["# one to one map\n","id_map1 = dict()\n","id_map2 = dict()\n","for i in range(len(node_ids)):\n"," id_map1[str(i)] = node_ids[i]\n"," id_map2[str(node_ids[i])] = i"],"metadata":{"id":"ULIoErBoPmzO","executionInfo":{"status":"ok","timestamp":1651555385318,"user_tz":-480,"elapsed":2,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":42,"outputs":[]},{"cell_type":"code","source":["# If we simply feed all pairs rudely into grah, there will be too many nodes which can cause session crash\n","# we decide to create a dictionary to store ids of nodes having at least one edge, and we redifine these nodes' ids\n","temp = 0\n","node_pairs = []\n","for x in range(len(file_data)//2-1):\n"," pair = file_data[x*2+1].strip('\\n').split(' ')[1:]\n"," if temp%50000==0:\n"," node_pairs = list(set(node_pairs))\n"," print(temp)\n"," temp = temp + 1\n"," l = len(pair)\n"," for i in range(l-1):\n"," for j in range(i+1, l-1):\n"," first, second = pair[i], pair[j]\n"," node_pairs.append((id_map2[first],id_map2[second]))"],"metadata":{"id":"Wtvh9yJRWMR5","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1651555387334,"user_tz":-480,"elapsed":2018,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"622496a4-74ef-41a2-d326-e84cf31be6c9"},"execution_count":43,"outputs":[{"output_type":"stream","name":"stdout","text":["0\n","50000\n"]}]},{"cell_type":"code","source":["node_pairs = list(set(node_pairs))"],"metadata":{"id":"WOlNBikRZMEq","executionInfo":{"status":"ok","timestamp":1651555387334,"user_tz":-480,"elapsed":6,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":44,"outputs":[]},{"cell_type":"code","source":["print(f\"# node pairs:{len(node_pairs)}\")\n","print(f\"# node ids:{len(node_ids)}\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"bqVKDRUCa4k8","executionInfo":{"status":"ok","timestamp":1651555387334,"user_tz":-480,"elapsed":5,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"27b92c55-051e-4538-a85e-4b68d4c8bff3"},"execution_count":45,"outputs":[{"output_type":"stream","name":"stdout","text":["# node pairs:746445\n","# node ids:75728\n"]}]},{"cell_type":"code","source":["g = Graph(node_pairs)"],"metadata":{"id":"IVsNbDpuSABP","executionInfo":{"status":"ok","timestamp":1651555387986,"user_tz":-480,"elapsed":5,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":46,"outputs":[]},{"cell_type":"code","source":["Degree = g.degree()"],"metadata":{"id":"Y9ZQzkrJSzcc","executionInfo":{"status":"ok","timestamp":1651555387986,"user_tz":-480,"elapsed":5,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":47,"outputs":[]},{"cell_type":"code","source":["for i in range(10):\n"," index = Degree.index(max(Degree))\n"," print(id_map1[str(index)], '|','|',Degree[index])\n"," Degree[index] = 0"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"4eXwbKKV9_7X","executionInfo":{"status":"ok","timestamp":1651555387986,"user_tz":-480,"elapsed":5,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"618dff17-dd5f-4ea6-a6ee-d790459a169f"},"execution_count":48,"outputs":[{"output_type":"stream","name":"stdout","text":["321278 | | 2306\n","27193779 | | 1808\n","206084 | | 1738\n","7691631 | | 1698\n","3228505 | | 1551\n","10270250 | | 1390\n","11061773 | | 1317\n","1420493 | | 1255\n","24560307 | | 1139\n","9384267 | | 1108\n"]}]},{"cell_type":"code","source":["import matplotlib.pyplot as plt\n","import numpy as np\n","plt.rcParams.update({'font.size': 25})\n","degrees = g.degree()\n","x = [x for x in range(max(degrees)+1)]\n","degree_counts = [0 for x in range(max(degrees)+1)]\n","\n","for i in degrees:\n"," degree_counts[i] += 1\n","\n","\n","plt.figure(figsize=(40,10))\n","plt.loglog(x, degree_counts, linewidth=3.0)\n","plt.ylabel('Number of vertices having the given degree')\n","plt.xlabel('Degree')\n","plt.title('Degree Distribution of Vertices in the CiteSeer Graph')\n","\n","plt.grid(True)\n","\n","plt.show()\n","plt.draw()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":363},"id":"0FwFBZzjCYDT","executionInfo":{"status":"ok","timestamp":1651555389850,"user_tz":-480,"elapsed":1867,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"7a694a79-7fa5-4dcf-c0e7-7a4934af5c52"},"execution_count":49,"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["
"]},"metadata":{}}]},{"cell_type":"code","source":["rank = g.pagerank(directed=False)"],"metadata":{"id":"z8hpn7ZoeE-t","executionInfo":{"status":"ok","timestamp":1651555389851,"user_tz":-480,"elapsed":9,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":50,"outputs":[]},{"cell_type":"code","source":["for i in range(10):\n"," index = rank.index(max(rank))\n"," print(id_map1[str(index)],'|','|', rank[index]*10000)\n"," rank[index] = 0"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"gx-nRA8NEbtG","executionInfo":{"status":"ok","timestamp":1651555389851,"user_tz":-480,"elapsed":8,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"2de173b8-83d8-4236-eef8-10df75c81f91"},"execution_count":51,"outputs":[{"output_type":"stream","name":"stdout","text":["321278 | | 12.836063163558814\n","206084 | | 12.182129136487386\n","7691631 | | 11.334998174029815\n","3228505 | | 9.581670921929092\n","27193779 | | 9.34194119126938\n","1420493 | | 8.09320255471226\n","10270250 | | 7.689296222718188\n","11061773 | | 6.591542444737357\n","19872456 | | 6.071249344805698\n","9384267 | | 5.9682403291913655\n"]}]},{"cell_type":"code","source":["degrees = g.degree()\n","rank = g.pagerank(directed=False)\n","final_de = []\n","final_ra = []\n","for i in range(10000):\n"," index = (rank.index(max(rank)))\n"," final_ra.append(id_map1[str(index)])\n"," rank[index] = 0\n"," index = (degrees.index(max(degrees)))\n"," final_de.append(id_map1[str(index)])\n"," degrees[index] = 0\n"],"metadata":{"id":"uXgp5T2GHKk_","executionInfo":{"status":"ok","timestamp":1651555433279,"user_tz":-480,"elapsed":43434,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":52,"outputs":[]},{"cell_type":"code","source":["import numpy as np\n","np.save(target+'_degree.npy', final_de)\n","np.save(target+'_rank.npy', final_ra)"],"metadata":{"id":"Vu6Kzwq1Hyvo","executionInfo":{"status":"ok","timestamp":1651555433280,"user_tz":-480,"elapsed":24,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":53,"outputs":[]},{"cell_type":"code","source":[""],"metadata":{"id":"Qo6ANanGIYPq","executionInfo":{"status":"ok","timestamp":1651555259074,"user_tz":-480,"elapsed":9,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":37,"outputs":[]}]} \ No newline at end of file diff --git a/notebooks/data_pull.ipynb b/notebooks/data_pull.ipynb new file mode 100644 index 0000000..dc963ca --- /dev/null +++ b/notebooks/data_pull.ipynb @@ -0,0 +1,306 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "odps.Schema {\n", + " login string \n", + " created_at date \n", + " database_id int64 \n", + " location string \n", + " company string \n", + " bio string \n", + " is_employee boolean \n", + " email string \n", + " infoname string \n", + " followers string \n", + " following string \n", + " time date \n", + " name string \n", + " lastupdatedat date \n", + " nextupdateat date \n", + "}\n", + "\n", + "odps.Schema {\n", + " id string \n", + " type string \n", + " action string \n", + " actor_id int64 \n", + " actor_login string \n", + " repo_id int64 \n", + " repo_name string \n", + " org_id int64 \n", + " org_login string \n", + " created_at datetime \n", + " issue_id int64 \n", + " issue_number int32 \n", + " issue_title string \n", + " issue_body string \n", + " issue_labels_name list \n", + " issue_labels_color list \n", + " issue_labels_default list \n", + " issue_labels_description list \n", + " issue_author_id int64 \n", + " issue_author_login string \n", + " issue_author_type string \n", + " issue_author_association string \n", + " issue_assignee_id int64 \n", + " issue_assignee_login string \n", + " issue_assignees_id list \n", + " issue_assignees_login list \n", + " issue_created_at datetime \n", + " issue_updated_at datetime \n", + " issue_comments int16 \n", + " issue_closed_at datetime \n", + " issue_comment_id int64 \n", + " issue_comment_body string \n", + " issue_comment_created_at datetime \n", + " issue_comment_updated_at datetime \n", + " issue_comment_author_association string \n", + " issue_comment_author_id int64 \n", + " issue_comment_author_login string \n", + " issue_comment_author_type string \n", + " pull_commits int16 \n", + " pull_additions int16 \n", + " pull_deletions int16 \n", + " pull_changed_files int32 \n", + " pull_merged int8 \n", + " pull_merge_commit_sha string \n", + " pull_merged_at datetime \n", + " pull_merged_by_id int64 \n", + " pull_merged_by_login string \n", + " pull_merged_by_type string \n", + " pull_requested_reviewer_id int64 \n", + " pull_requested_reviewer_login string \n", + " pull_requested_reviewer_type string \n", + " pull_review_comments int16 \n", + " repo_description string \n", + " repo_size int32 \n", + " repo_stargazers_count int32 \n", + " repo_forks_count int32 \n", + " repo_language string \n", + " repo_has_issues int8 \n", + " repo_has_projects int8 \n", + " repo_has_downloads int8 \n", + " repo_has_wiki int8 \n", + " repo_has_pages int8 \n", + " repo_license string \n", + " repo_default_branch string \n", + " repo_created_at datetime \n", + " repo_updated_at datetime \n", + " repo_pushed_at datetime \n", + " pull_review_id int64 \n", + " pull_review_comment_id int64 \n", + " pull_review_comment_path string \n", + " pull_review_comment_position string \n", + " pull_review_comment_author_id int64 \n", + " pull_review_comment_author_login string \n", + " pull_review_comment_author_type string \n", + " pull_review_comment_author_association string \n", + " pull_review_comment_body string \n", + " pull_review_comment_created_at datetime \n", + " pull_review_comment_updated_at datetime \n", + " push_id int64 \n", + " push_size int32 \n", + " push_distinct_size int32 \n", + " push_ref string \n", + " push_head string \n", + " push_before string \n", + " push_commits_name list \n", + " push_commits_email list \n", + " push_commits_message list \n", + " fork_forkee_id int64 \n", + " fork_forkee_full_name string \n", + " fork_forkee_owner_id int64 \n", + " fork_forkee_owner_login string \n", + " fork_forkee_owner_type string \n", + " delete_ref string \n", + " delete_ref_type string \n", + " delete_pusher_type string \n", + " create_ref string \n", + " create_ref_type string \n", + " create_master_branch string \n", + " create_description string \n", + " create_pusher_type string \n", + " gollum_pages_page_name list \n", + " gollum_pages_title list \n", + " gollum_pages_action list \n", + " member_login string \n", + " member_type string \n", + " member_id int64 \n", + " release_id int64 \n", + " release_tag_name string \n", + " release_target_commitish string \n", + " release_name string \n", + " release_draft int8 \n", + " release_author_id int64 \n", + " release_author_login string \n", + " release_author_type string \n", + " release_prerelease int8 \n", + " release_created_at datetime \n", + " release_published_at datetime \n", + " release_body string \n", + " release_assets_name list \n", + " release_assets_uploader_login list \n", + " release_assets_uploader_id list \n", + " release_assets_content_type list \n", + " release_assets_state list \n", + " release_assets_size list \n", + " release_assets_download_count list \n", + " commit_comment_id int64 \n", + " commit_comment_author_id int64 \n", + " commit_comment_author_login string \n", + " commit_comment_author_type string \n", + " commit_comment_author_association string \n", + " commit_comment_body string \n", + " commit_comment_path string \n", + " commit_comment_position string \n", + " commit_comment_line string \n", + " commit_comment_created_at datetime \n", + " commit_comment_updated_at datetime \n", + " pt string \n", + "}\n", + "\n" + ] + } + ], + "source": [ + "from odps import ODPS\n", + "from odps import options\n", + "from odps.df import DataFrame\n", + "import pandas as pd\n", + "import numpy as np\n", + "\n", + "pd.set_option('display.max_rows',None)\n", + "\n", + "ACCESS_ID = 'LTAI5t9uwJrh5eJ7Q5E37D1s'\n", + "SECRET_ACCESS_KEY = 'NCFHOAnvqfnTrpypgR4b3cNawP8fnB'\n", + "ODPS_PROJECT = 'OpenDigger_prod_dev'\n", + "ODPS_ENDPOINT = 'http://service.cn-shanghai.maxcompute.aliyun.com/api'\n", + "\n", + "o = ODPS(ACCESS_ID, SECRET_ACCESS_KEY,\n", + " project=ODPS_PROJECT, endpoint=ODPS_ENDPOINT)\n", + "options.tunnel.limit_instance_tunnel = False\n", + "# options.read_timeout = 10000000\n", + "\n", + "users = DataFrame(o.get_table('ods_github_users'))\n", + "print(users.dtypes)\n", + "\n", + "github_log = DataFrame(o.get_table('ods_github_log'))\n", + "print(github_log.dtypes)" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [], + "source": [ + "sql = '''\n", + " select type, count(repo_id), repo_id from ods_github_log\n", + " where pt='20151001'\n", + " and type in ('PullRequestEvent','WatchEvent','ForkEvent','IssueCommentEvent')\n", + " group by type, repo_id;\n", + "'''\n", + "\n", + "result = o.execute_sql(sql, hints={'odps.sql.allow.fullscan': 'true', 'odps.sql.submit.mode': 'script'})\n", + "with open('data\\count.txt', 'w') as f:\n", + " with result.open_reader() as reader:\n", + " for record in reader:\n", + " type = record['type']\n", + " count = record['_c1']\n", + " repo_id = record['repo_id'] \n", + " f.write('type: {type}, repo_id: {repo_id}, count: {count}\\n'.format(\n", + " type=type,\n", + " repo_id=repo_id,\n", + " count=count)) " + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [], + "source": [ + "graph_dict = {}\n", + "graph_dict['PullRequestEvent'] = {}\n", + "graph_dict['WatchEvent'] = {}\n", + "graph_dict['ForkEvent'] = {}\n", + "graph_dict['IssueCommentEvent'] = {}\n", + "sql = '''\n", + " select type, repo_id, actor_id\n", + " from ods_github_log\n", + " where pt='20151001'\n", + " and type in ('PullRequestEvent','WatchEvent','ForkEvent','IssueCommentEvent')\n", + " group by type, repo_id, actor_id;\n", + "'''\n", + "result = o.execute_sql(sql, hints={'odps.sql.allow.fullscan': 'true', 'odps.sql.submit.mode': 'script'})\n", + "with result.open_reader() as reader:\n", + " for record in reader:\n", + " type = record['type']\n", + " actor_id = record['actor_id']\n", + " repo_id = record['repo_id']\n", + " if actor_id not in graph_dict[type]:\n", + " graph_dict[type][actor_id] = []\n", + " graph_dict[type][actor_id].append(str(repo_id))\n", + "\n", + "# print(graph_dict)\n", + "\n", + "with open('data\\PullRequestEvent.txt', 'w') as f:\n", + " for key in graph_dict['PullRequestEvent']:\n", + " if len(graph_dict['PullRequestEvent'][key]) < 2:\n", + " continue\n", + " f.write('actor_id: {actor_id}\\nrepo_id: {list}\\n'.format(actor_id=key, list=' '.join(graph_dict['PullRequestEvent'][key])))\n", + "\n", + "with open('data\\WatchEvent.txt', 'w') as f:\n", + " for key in graph_dict['WatchEvent']:\n", + " if len(graph_dict['WatchEvent'][key]) < 2:\n", + " continue\n", + " f.write('actor_id: {actor_id}\\nrepo_id: {list}\\n'.format(actor_id=key, list=' '.join(graph_dict['WatchEvent'][key])))\n", + "\n", + "with open('data\\ForkEvent.txt', 'w') as f:\n", + " for key in graph_dict['ForkEvent']:\n", + " if len(graph_dict['ForkEvent'][key]) < 2:\n", + " continue\n", + " f.write('actor_id: {actor_id}\\nrepo_id: {list}\\n'.format(actor_id=key, list=' '.join(graph_dict['ForkEvent'][key])))\n", + "\n", + "with open('data\\IssueCommentEvent.txt', 'w') as f:\n", + " for key in graph_dict['IssueCommentEvent']:\n", + " if len(graph_dict['IssueCommentEvent'][key]) < 2:\n", + " continue\n", + " f.write('actor_id: {actor_id}\\nrepo_id: {list}\\n'.format(actor_id=key, list=' '.join(graph_dict['IssueCommentEvent'][key])))" + ] + } + ], + "metadata": { + "interpreter": { + "hash": "caac794e4b8e34bcc9a4d9e1a06492e263031294735d822cbf2db7854bb6c6da" + }, + "kernelspec": { + "display_name": "Python 3.10.4 64-bit (windows store)", + "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": "3.10.4" + }, + "orig_nbformat": 4 + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/notebooks/learning.ipynb b/notebooks/learning.ipynb new file mode 100644 index 0000000..78f798d --- /dev/null +++ b/notebooks/learning.ipynb @@ -0,0 +1 @@ +{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"learning.ipynb","provenance":[],"authorship_tag":"ABX9TyOH0O9m/wI7scpeov7qGokX"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"code","execution_count":43,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"cMHw9PfA4PdG","executionInfo":{"status":"ok","timestamp":1651650933526,"user_tz":-480,"elapsed":3253,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"c546da43-6e70-4e0e-f90d-c167e3bc5cfc"},"outputs":[{"output_type":"stream","name":"stdout","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"]}],"source":["from google.colab import drive\n","drive.mount('/content/drive')\n","import os\n","import torch\n","import random"]},{"cell_type":"code","source":["path = '/content/drive/My Drive/social_computing/data/count.txt'"],"metadata":{"id":"in4a4Lmr4R5h","executionInfo":{"status":"ok","timestamp":1651650136175,"user_tz":-480,"elapsed":2,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":30,"outputs":[]},{"cell_type":"code","source":["f = open(path, encoding = \"utf-8\")\n","file_data = f.readlines()"],"metadata":{"id":"YynmL5fP4bwT","executionInfo":{"status":"ok","timestamp":1651650136175,"user_tz":-480,"elapsed":2,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":31,"outputs":[]},{"cell_type":"code","source":["from typing import Dict\n","id_list = []\n","for line in file_data:\n"," id_list.append(line.strip('\\n').split(' ')[3][:-1])\n","\n","# data_list['id'] = [PullRequest_count, Fork_count, Watch_count, IssueComment_count, class]\n","map = dict()\n","map['PullRequestEvent'] = 0\n","map['ForkEvent'] = 1\n","map['WatchEvent'] = 2\n","map['IssueCommentEvent'] = 3\n","data_list = dict()"],"metadata":{"id":"Uz02esuk6bRl","executionInfo":{"status":"ok","timestamp":1651650137149,"user_tz":-480,"elapsed":976,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":32,"outputs":[]},{"cell_type":"code","source":["# Initialize data_list\n","for id in id_list:\n"," data_list[id] = [0,0,0,0,0]"],"metadata":{"id":"JuyvyC-l9gRU","executionInfo":{"status":"ok","timestamp":1651650139027,"user_tz":-480,"elapsed":1885,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":33,"outputs":[]},{"cell_type":"code","source":["for line in file_data: \n"," raw = line.strip('\\n').split(' ') \n"," data_list[raw[3][:-1]][map[raw[1][:-1]]] = int(raw[-1])"],"metadata":{"id":"72-EFogQ82I2","executionInfo":{"status":"ok","timestamp":1651650140722,"user_tz":-480,"elapsed":1697,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":34,"outputs":[]},{"cell_type":"code","source":["p_list = []\n","final_data = []\n","for id in id_list:\n"," p_list.append(data_list[id][0])\n","p_list.sort(reverse=True)\n","threshold = p_list[int(len(p_list)*0.1)]\n","for id in id_list:\n"," if data_list[id][0]>=threshold:\n"," data_list[id][4] = 1\n"," final_data.append(data_list[id][1:])"],"metadata":{"id":"OiHpD_rX4nFS","executionInfo":{"status":"ok","timestamp":1651650364944,"user_tz":-480,"elapsed":3201,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":38,"outputs":[]},{"cell_type":"code","source":["final_data = torch.tensor(final_data)"],"metadata":{"id":"W7d1zzzaCSVS","executionInfo":{"status":"ok","timestamp":1651650424843,"user_tz":-480,"elapsed":473,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":39,"outputs":[]},{"cell_type":"code","source":["print(final_data.shape)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"VAQlRswJCXEy","executionInfo":{"status":"ok","timestamp":1651650434749,"user_tz":-480,"elapsed":409,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"5e2bf89e-ec92-4149-aa8b-01b42f232332"},"execution_count":40,"outputs":[{"output_type":"stream","name":"stdout","text":["torch.Size([883436, 4])\n"]}]},{"cell_type":"code","source":["class Dataset(torch.utils.data.Dataset):\n"," def __init__(self, data):\n"," self.labels = data[:, -1]\n"," self.x = data[:, 0:-1]\n"," def __len__(self):\n"," return len(self.labels)\n","\n"," def __getitem__(self, idx):\n"," x = self.x[idx].float()\n"," y = self.labels[idx]\n"," return x, y"],"metadata":{"id":"gPygHvtPBDZO","executionInfo":{"status":"ok","timestamp":1651651824344,"user_tz":-480,"elapsed":490,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":68,"outputs":[]},{"cell_type":"code","source":["# randomly sample 70% as training data, and remaining 30% data as testing data\n","sample = [i for i in range(len(final_data))]\n","sample = random.sample(sample, len(final_data))\n","train = Dataset(final_data[sample[:int(len(final_data)*0.7)]])\n","train_dataloader = torch.utils.data.DataLoader(train, batch_size=256, shuffle=True, drop_last=True)"],"metadata":{"id":"McymePR1CtOl","executionInfo":{"status":"ok","timestamp":1651651827418,"user_tz":-480,"elapsed":1657,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":69,"outputs":[]},{"cell_type":"code","source":["from torch.optim import Adam\n","from tqdm import tqdm\n","from torch import nn\n","def train(model, trainloader, learning_rate, epochs):\n"," use_cuda = torch.cuda.is_available()\n"," device = torch.device(\"cuda\" if use_cuda else \"cpu\")\n"," criterion = nn.CrossEntropyLoss()\n"," optimizer = Adam(model.parameters(), lr= learning_rate)\n"," if use_cuda:\n"," model = model.cuda()\n"," criterion = criterion.cuda()\n"," for epoch_num in range(epochs):\n"," acc_train = 0\n"," loss_train = 0\n"," for train_input, train_label in tqdm(trainloader):\n"," model.zero_grad()\n"," train_input = train_input.to(device)\n"," train_label = train_label.to(device)\n"," output = model(train_input)\n"," batch_loss = criterion(output, train_label)\n"," loss_train += batch_loss.item()\n"," batch_loss.backward()\n"," optimizer.step()\n"," with torch.no_grad():\n"," label_new = output.argmax(dim=1)\n"," acc_train = acc_train + (train_label==label_new).float().mean()\n"," with torch.no_grad():\n"," loss_train = loss_train / len(trainloader)\n"," acc_train = acc_train / len(trainloader)\n"," print(f'training loss:{loss_train}, training accuracy:{acc_train}')"],"metadata":{"id":"piLHSHz2EQ4J","executionInfo":{"status":"ok","timestamp":1651651827418,"user_tz":-480,"elapsed":3,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":70,"outputs":[]},{"cell_type":"code","source":["model = nn.Sequential(nn.Linear(3,2))\n","train(model, trainloader=train_dataloader, learning_rate=11, epochs=20)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":346},"id":"4SfwM6QhGggu","executionInfo":{"status":"error","timestamp":1651652161257,"user_tz":-480,"elapsed":17598,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"73e9fef0-0b46-4e36-8a2e-2e240723dfcf"},"execution_count":78,"outputs":[{"output_type":"stream","name":"stderr","text":["100%|██████████| 4831/4831 [00:10<00:00, 465.96it/s]\n"]},{"output_type":"stream","name":"stdout","text":["training loss:68.62970850901237, training accuracy:0.8554679155349731\n"]},{"output_type":"stream","name":"stderr","text":[" 64%|██████▍ | 3091/4831 [00:06<00:03, 442.70it/s]\n"]},{"output_type":"error","ename":"KeyboardInterrupt","evalue":"ignored","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)","\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mSequential\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mLinear\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m2\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----> 2\u001b[0;31m \u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrainloader\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtrain_dataloader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlearning_rate\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m11\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m20\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m","\u001b[0;32m\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(model, trainloader, learning_rate, epochs)\u001b[0m\n\u001b[1;32m 13\u001b[0m \u001b[0macc_train\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 14\u001b[0m \u001b[0mloss_train\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 15\u001b[0;31m \u001b[0;32mfor\u001b[0m \u001b[0mtrain_input\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrain_label\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mtqdm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrainloader\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 16\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzero_grad\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 17\u001b[0m \u001b[0mtrain_input\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_input\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.7/dist-packages/tqdm/std.py\u001b[0m in \u001b[0;36m__iter__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 1193\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1194\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1195\u001b[0;31m \u001b[0;32mfor\u001b[0m \u001b[0mobj\u001b[0m \u001b[0;32min\u001b[0m \u001b[0miterable\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 1196\u001b[0m \u001b[0;32myield\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1197\u001b[0m \u001b[0;31m# Update and possibly print the progressbar.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py\u001b[0m in \u001b[0;36m__next__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 528\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sampler_iter\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 529\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_reset\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--> 530\u001b[0;31m \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_next_data\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 531\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_num_yielded\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 532\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dataset_kind\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0m_DatasetKind\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIterable\u001b[0m \u001b[0;32mand\u001b[0m\u001b[0;31m \u001b[0m\u001b[0;31m\\\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py\u001b[0m in \u001b[0;36m_next_data\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 568\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_next_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\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 569\u001b[0m \u001b[0mindex\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_next_index\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# may raise StopIteration\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 570\u001b[0;31m \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dataset_fetcher\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfetch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindex\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# may raise StopIteration\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 571\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_pin_memory\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 572\u001b[0m \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_utils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpin_memory\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpin_memory\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/usr/local/lib/python3.7/dist-packages/torch/utils/data/_utils/fetch.py\u001b[0m in \u001b[0;36mfetch\u001b[0;34m(self, possibly_batched_index)\u001b[0m\n\u001b[1;32m 50\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 51\u001b[0m \u001b[0mdata\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[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 52\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcollate_fn\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[0;31mKeyboardInterrupt\u001b[0m: "]}]},{"cell_type":"code","source":["test_data = final_data[sample[int(len(final_data)*0.7):]][:, :-1].float()\n","test_label = final_data[sample[int(len(final_data)*0.7):]][:, -1].long()\n","(model(test_data).argmax(dim=1)==test_label).float().mean()"],"metadata":{"id":"P6D3S8uaIZch","executionInfo":{"status":"ok","timestamp":1651652046350,"user_tz":-480,"elapsed":406,"user":{"displayName":"ming li","userId":"14148720490428311514"}}},"execution_count":76,"outputs":[]},{"cell_type":"code","source":[""],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"2xjuln6OGqoM","executionInfo":{"status":"ok","timestamp":1651652080979,"user_tz":-480,"elapsed":391,"user":{"displayName":"ming li","userId":"14148720490428311514"}},"outputId":"af0457f4-ec9f-4c63-9f16-55a421cb3a08"},"execution_count":77,"outputs":[{"output_type":"execute_result","data":{"text/plain":["tensor(0.9002)"]},"metadata":{},"execution_count":77}]},{"cell_type":"code","source":[""],"metadata":{"id":"8Mj19ZN-IrH2"},"execution_count":null,"outputs":[]}]} \ No newline at end of file