Skip to content

Pull Requests #42

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open

Pull Requests #42

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
49d38c4
add stackexchange code
loubnabnl Sep 20, 2023
58dde17
add code
loubnabnl Sep 20, 2023
3ef2a86
add notebook
loubnabnl Sep 20, 2023
e670afa
add notebook
loubnabnl Sep 21, 2023
df0dc1b
update
loubnabnl Sep 21, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
349 changes: 349 additions & 0 deletions data_analysis/pull-requests/reconstruct_prs.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,349 @@
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Analysis"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Requirement already satisfied: python-dateutil in /Users/loubnabenallal/Desktop/HF_bigcode/.venv/lib/python3.11/site-packages (2.8.2)\n",
"Requirement already satisfied: six>=1.5 in /Users/loubnabenallal/Desktop/HF_bigcode/.venv/lib/python3.11/site-packages (from python-dateutil) (1.16.0)\n",
"\n",
"\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.0.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m23.2.1\u001b[0m\n",
"\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n",
"Note: you may need to restart the kernel to use updated packages.\n"
]
}
],
"source": [
"pip install python-dateutil"
]
},
{
"cell_type": "code",
"execution_count": 330,
"metadata": {},
"outputs": [],
"source": [
"import json\n",
"import pandas as pd\n",
"from dateutil.parser import parse\n",
"from datasets import load_dataset, Dataset\n",
"\n",
"small_ds = load_dataset(\"bigcode-data/the-stack-gh-pull-requests\", use_auth_token=True, split=\"train\", streaming=True)\n",
"\n",
"size = 500_000\n",
"\n",
"ds = small_ds.shuffle(seed=0, buffer_size=1_000_000)\n",
"\n",
"# 10k subset of random samples from ds\n",
"ds = list(ds.take(size))\n",
"ds = Dataset.from_pandas(pd.DataFrame(data=ds))"
]
},
{
"cell_type": "code",
"execution_count": 365,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Dataset({\n",
" features: ['pull_request.guid', 'pull_request.code_review_events', 'pull_request.events', 'pull_request.issue_events', 'bucket', '__index_level_0__'],\n",
" num_rows: 500000\n",
"})"
]
},
"execution_count": 365,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ds"
]
},
{
"cell_type": "code",
"execution_count": 444,
"metadata": {},
"outputs": [],
"source": [
"# merge all three instances\n",
"\n",
"pull_request_info_cols = [\n",
" \"repo.name\",\n",
" \"repo.id\",\n",
" \"org.id\",\n",
" \"public\",\n",
" \"pull_request.id\",\n",
" \"pull_request.guid\",\n",
" \"pull_request.number\",\n",
" \"pull_request.title\",\n",
" \"pull_request.body\",\n",
" \"pull_request.state\",\n",
" \"pull_request.user.login\",\n",
" \"pull_request.user.id\",\n",
" # add user type\n",
" \"pull_request.head.user.type\",\n",
" \"pull_request.base.user.type\",\n",
" \"pull_request.created_at\",\n",
" \"pull_request.closed_at\",\n",
" \"pull_request.merged_at\",\n",
" \"pull_request.merged_by.login\",\n",
" \"pull_request.milestone.title\",\n",
" \"pull_request.milestone.description\",\n",
" \"pull_request.milestone.number\",\n",
" # commits\n",
" 'pull_request.commits',\n",
" 'pull_request.additions',\n",
" 'pull_request.deletions',\n",
" # changed files\n",
" 'pull_request.changed_files',\n",
" \"pull_request.comments\",\n",
" \"pull_request.review_comments\",\n",
"]\n",
"\n",
"head_info_cols = [\n",
" \"pull_request.head.label\",\n",
" \"pull_request.head.ref\",\n",
" \"pull_request.head.user.login\",\n",
" \"pull_request.head.user.type\",\n",
" \"pull_request.head.repo.owner.login\",\n",
" \"pull_request.head.repo.owner.type\",\n",
" \"pull_request.head.repo.license.name\",\n",
" \"pull_request.head.sha\",\n",
" 'pull_request.head.repo.name',\n",
" 'pull_request.head.repo.owner.login',\n",
" 'pull_request.head.repo.homepage',\n",
" 'pull_request.head.repo.description',\n",
" 'pull_request.head.repo.language',\n",
" 'pull_request.head.repo.stargazers_count',\n",
" 'pull_request.head.repo.license.name',\n",
" 'pull_request.head.repo.default_branch',\n",
" 'pull_request.head.repo.private'\n",
"]\n",
"base_info_cols = [\n",
" \"pull_request.base.label\",\n",
" \"pull_request.base.ref\",\n",
" \"pull_request.base.sha\",\n",
" \"pull_request.base.user.login\",\n",
" \"pull_request.base.user.type\",\n",
" \"pull_request.base.repo.owner.login\",\n",
" \"pull_request.base.repo.owner.type\",\n",
" \"pull_request.base.repo.license.name\",\n",
" \"pull_request.base.repo.default_branch\",\n",
" \"pull_request.base.repo.description\",\n",
" \"pull_request.base.repo.language\",\n",
" \"pull_request.base.repo.watchers_count\",\n",
" \"pull_request.base.repo.open_issues_count\",\n",
" \"pull_request.base.repo.forks_count\",\n",
" 'pull_request.base.repo.name',\n",
" 'pull_request.base.repo.owner.login',\n",
" 'pull_request.base.repo.homepage',\n",
" 'pull_request.base.repo.description',\n",
" 'pull_request.base.repo.language',\n",
" 'pull_request.base.repo.stargazers_count',\n",
" 'pull_request.base.repo.private',\n",
" 'pull_request.comments',\n",
" 'pull_request.review_comments',\n",
" 'pull_request.label.name',\n",
"]\n",
"\n",
"reviews_info = [# review events only\n",
" 'actor.login',\n",
" 'actor.id',\n",
" 'user.login',\n",
" 'user.type',\n",
" 'review.state',\n",
" 'review.id', \n",
" 'review.body', \n",
" 'review.commit_id', \n",
" 'review.submitted_at', \n",
" 'review.author_association',\n",
" \"pull_request.state\",\n",
" \"pull_request.merged\",\n",
" \"pull_request.merged_by.login\",\n",
" \"pull_request.merged_by.type\",\n",
" # comments\n",
" 'comment.id',\n",
" 'comment.diff_hunk',\n",
" 'comment.body',\n",
" 'comment.path',\n",
" 'comment.position',\n",
" 'comment.original_position',\n",
" 'comment.commit_id',\n",
" 'comment.original_commit_id',\n",
" 'comment.created_at',\n",
" 'comment.updated_at',\n",
" 'comment.author_association',\n",
" 'comment.start_line',\n",
" 'comment.original_start_line',\n",
" 'comment.start_side',\n",
" 'comment.line',\n",
" 'comment.original_line',\n",
" 'comment.side',\n",
" 'comment.in_reply_to_id',]\n",
"\n",
"\n",
"issues_info = [\n",
" 'author',\n",
" 'comment',\n",
" 'comment_id']\n",
" \n",
"event_info = reviews_info + issues_info\n",
"\n",
"def get_event_info(review):\n",
" res = {k: review[k] if k in review else None for k in event_info}\n",
" # for keys in issues_info add prefix issue.\n",
" for k in issues_info:\n",
" res[\"issue.\" + k] = res[k]\n",
" del res[k]\n",
" return res\n",
"\n",
"def load_json(data):\n",
" try:\n",
" data = json.loads(data)\n",
" if isinstance(data, dict):\n",
" data = [data]\n",
" return data\n",
" except TypeError:\n",
" return []\n",
"\n",
"def update_datetime(e):\n",
" e[\"created_at\"] = parse(e[\"created_at\"])\n",
" return e\n",
"\n",
"def merge_events(row):\n",
" events = load_json(row[\"pull_request.events\"])\n",
" reviews = load_json(row[\"pull_request.code_review_events\"])\n",
" issues = load_json(row[\"pull_request.issue_events\"])\n",
"\n",
" assert len(issues) <= 1\n",
" if issues:\n",
" issues_events = issues[0][\"events\"]\n",
" # for each events in each category group all events sorted by \"created_at\" in one list\n",
" for e in issues_events:\n",
" e[\"created_at\"] = parse(e[\"datetime\"])\n",
" del e[\"datetime\"]\n",
" else:\n",
" issues_events = []\n",
" events = [update_datetime(e) for e in events]\n",
" reviews = [update_datetime(e) for e in reviews]\n",
" all_events = sorted(\n",
" events + reviews + issues_events,\n",
" key=lambda x: x[\"created_at\"]\n",
" )\n",
" try:\n",
" base_data = events[0] if events else reviews[0]\n",
" except IndexError:\n",
" # init empty dict\n",
" base_data = {}\n",
" if issues:\n",
" base_data = {}\n",
" first_event = issues[0]\n",
" base_data['pull_request.title'] = first_event[\"events\"][0][\"title\"]\n",
" base_data[\"repo.name\"] = first_event[\"repo\"]\n",
" base_data[\"pull_request.number\"] = first_event[\"pull_request\"][\"number\"]\n",
" base_data[\"pull_request.user.login\"] = first_event[\"pull_request\"][\"user_login\"]\n",
" print(\"filling PR data from issue event\")\n",
" else:\n",
" raise IndexError(\"No events for PR\")\n",
" \n",
" # Initialize with default values\n",
" pr_info = {k: None for k in pull_request_info_cols}\n",
" head_info = {k: None for k in head_info_cols}\n",
" base_info = {k: None for k in base_info_cols}\n",
"\n",
" # Fill available data\n",
" pr_info.update({k: base_data[k] if k in base_data else None for k in pull_request_info_cols})\n",
" head_info.update({k: base_data[k] if k in base_data else None for k in head_info_cols })\n",
" base_info.update({k: base_data[k] if k in base_data else None for k in base_info_cols})\n",
"\n",
" # each comment should have \"comments\" and \"review_comments\" fields with \"extra_review_info\" field\n",
" comments = [{\"type\": e[\"type\"],\n",
" \"action\": e[\"action\"],\n",
" \"created_at\": e[\"created_at\"],\n",
" **get_event_info(e)} for e in all_events]\n",
" new_row = {\"pull_request_info\": pr_info, \"head_repo_info\": head_info, \"base_repo_info\": base_info, \"events\": comments}\n",
" return new_row"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"merged_ds = ds.map(merge_events, remove_columns=[\"pull_request.events\", \"pull_request.code_review_events\", \"pull_request.issue_events\", '__index_level_0__','pull_request.guid'])"
]
},
{
"cell_type": "code",
"execution_count": 449,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Creating parquet from Arrow format: 100%|██████████| 84/84 [00:02<00:00, 37.20ba/s]\n",
"Upload 1 LFS files: 100%|██████████| 1/1 [00:22<00:00, 22.10s/it]\n",
"Creating parquet from Arrow format: 100%|██████████| 84/84 [00:02<00:00, 33.12ba/s]s/it]\n",
"Upload 1 LFS files: 100%|██████████| 1/1 [00:09<00:00, 9.55s/it]\n",
"Creating parquet from Arrow format: 100%|██████████| 84/84 [00:02<00:00, 39.47ba/s]s/it]\n",
"Upload 1 LFS files: 100%|██████████| 1/1 [00:09<00:00, 9.99s/it]\n",
"Creating parquet from Arrow format: 100%|██████████| 84/84 [00:02<00:00, 37.45ba/s]s/it]\n",
"Upload 1 LFS files: 100%|██████████| 1/1 [00:23<00:00, 23.74s/it]\n",
"Creating parquet from Arrow format: 100%|██████████| 84/84 [00:02<00:00, 34.84ba/s]s/it]\n",
"Upload 1 LFS files: 100%|██████████| 1/1 [00:22<00:00, 22.48s/it]\n",
"Creating parquet from Arrow format: 100%|██████████| 84/84 [00:03<00:00, 26.04ba/s]s/it]\n",
"Upload 1 LFS files: 100%|██████████| 1/1 [00:22<00:00, 22.62s/it]\n",
"Pushing dataset shards to the dataset hub: 100%|██████████| 6/6 [02:10<00:00, 21.69s/it]\n"
]
}
],
"source": [
"merged_ds.push_to_hub(\"loubnabnl/code_reviews_500k\")"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3.11.0 ('.venv': venv)",
"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": "3.11.0"
},
"orig_nbformat": 4,
"vscode": {
"interpreter": {
"hash": "0cc3054246fa39b40b564a97820c10836c9fb6acdf94e9196ea3a787cac26526"
}
}
},
"nbformat": 4,
"nbformat_minor": 2
}
3 changes: 3 additions & 0 deletions data_analysis/stackoverflow/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
## Code for processing StackExchange data

Code for processing stackexchange data dump available in `h4_code` (to build https://huggingface.co/datasets/HuggingFaceH4/stack-exchange-preferences) and `other`, notebook for further processing (e.g convert all HTML to Markdown) in `StackExchangeProcessing.ipynb` (to build https://huggingface.co/datasets/lvwerra/stack-exchange-paired)
Loading