Dataset Viewer (First 5GB)
Auto-converted to Parquet Duplicate
text
stringlengths
232
16.3k
domain
stringclasses
1 value
difficulty
stringclasses
3 values
meta
dict
<|fim_suffix|> :param obj: input object :return: Serializer """ # 1-NULL serializer if obj is None: return self._null_serializer_adapter obj_type = type(obj) serializer = None # 2-Default serializers, Dataserializable, Portable, primitives, ar...
code_fim
hard
{ "lang": "python", "repo": "mustafaiman/hazelcast-python-client", "path": "/hazelcast/serialization/base.py", "mode": "spm", "license": "Apache-2.0", "source": "the-stack-v2" }
<|fim_prefix|># repo: mustafaiman/hazelcast-python-client path: /hazelcast/serialization/base.py from threading import RLock from api import * from data import * from hazelcast.core import * from serializer import * EMPTY_PARTITIONING_STRATEGY = lambda key: None def handle_exception(e): if isinstance(e, Memory...
code_fim
hard
{ "lang": "python", "repo": "mustafaiman/hazelcast-python-client", "path": "/hazelcast/serialization/base.py", "mode": "psm", "license": "Apache-2.0", "source": "the-stack-v2" }
<|fim_suffix|> serializer_adaptor = create_buffer_serializer_wrapper(serializer) self.register_constant_serializer_adaptor(obj_type, serializer_adaptor) def register_constant_serializer_adaptor(self, obj_type, serializer_adaptor): self._constant_type_dict[obj_type] = serializer_adaptor ...
code_fim
hard
{ "lang": "python", "repo": "mustafaiman/hazelcast-python-client", "path": "/hazelcast/serialization/base.py", "mode": "spm", "license": "Apache-2.0", "source": "the-stack-v2" }
<|fim_suffix|> """Logs epoch level validation metrics.""" self.log_dict(self.val_metrics.compute()) self.val_metrics.reset() def test_step(self, *args: Any, **kwargs: Any) -> None: """Compute test loss. Args: batch: the output of your DataLoader """ ...
code_fim
hard
{ "lang": "python", "repo": "microsoft/torchgeo", "path": "/torchgeo/trainers/segmentation.py", "mode": "spm", "license": "MIT", "source": "the-stack-v2" }
<|fim_suffix|> """Logs epoch level test metrics.""" self.log_dict(self.test_metrics.compute()) self.test_metrics.reset() def predict_step(self, *args: Any, **kwargs: Any) -> Tensor: """Compute and return the predictions. By default, this will loop over images in a dataloader...
code_fim
hard
{ "lang": "python", "repo": "microsoft/torchgeo", "path": "/torchgeo/trainers/segmentation.py", "mode": "spm", "license": "MIT", "source": "the-stack-v2" }
<|fim_prefix|># repo: microsoft/torchgeo path: /torchgeo/trainers/segmentation.py # Copyright (c) Microsoft Corporation. All rights reserved. # Licensed under the MIT License. """Segmentation tasks.""" import os import warnings from typing import Any, cast import matplotlib.pyplot as plt import segmentation_models_...
code_fim
hard
{ "lang": "python", "repo": "microsoft/torchgeo", "path": "/torchgeo/trainers/segmentation.py", "mode": "psm", "license": "MIT", "source": "the-stack-v2" }
<|fim_prefix|># repo: vmware/network-insight-sdk-generic-datasources path: /network_insight_sdk_generic_datasources/common/import_module_utilities.py # Copyright 2019 VMware, Inc. # SPDX-License-Identifier: BSD-2-Clause import importlib def load_class(class_path): module = importlib.import_module(".".join(clas...
code_fim
hard
{ "lang": "python", "repo": "vmware/network-insight-sdk-generic-datasources", "path": "/network_insight_sdk_generic_datasources/common/import_module_utilities.py", "mode": "psm", "license": "BSD-2-Clause", "source": "the-stack-v2" }
<|fim_suffix|> def load_class_for_pre_post_parser(device, class_name): module_path = "{0}.{1}.{1}_{2}".format("network_insight_sdk_generic_datasources.routers_and_switches", device, "pre_post_processor") module = importlib.import_module(module_path) return getattr(...
code_fim
hard
{ "lang": "python", "repo": "vmware/network-insight-sdk-generic-datasources", "path": "/network_insight_sdk_generic_datasources/common/import_module_utilities.py", "mode": "spm", "license": "BSD-2-Clause", "source": "the-stack-v2" }
<|fim_suffix|>def load_class_for_process_table(device, class_name): module_path = "{0}.{1}.{1}_{2}".format("network_insight_sdk_generic_datasources.routers_and_switches", device, "pre_post_processor") module = importlib.import_module(module_path) return getattr(mod...
code_fim
hard
{ "lang": "python", "repo": "vmware/network-insight-sdk-generic-datasources", "path": "/network_insight_sdk_generic_datasources/common/import_module_utilities.py", "mode": "spm", "license": "BSD-2-Clause", "source": "the-stack-v2" }
<|fim_prefix|># repo: tnakamura/SilverTask path: /silvertask-web/kay/sessions/middleware.py # -*- coding: utf-8 -*- """ Kay session middleware. :Copyright: (c) 2009 Accense Technology, Inc. All rights reserved. :license: BSD, see LICENSE for more details. """ import kay.sessions from kay.conf import settings from w...
code_fim
medium
{ "lang": "python", "repo": "tnakamura/SilverTask", "path": "/silvertask-web/kay/sessions/middleware.py", "mode": "psm", "license": "MIT", "source": "the-stack-v2" }
<|fim_suffix|> if hasattr(request, '_cached_session') and \ request.session.should_save and hasattr(response, 'set_cookie'): session_store = import_string(settings.SESSION_STORE)() session_store.save(request.session) response.set_cookie(settings.COOKIE_NAME, sess...
code_fim
hard
{ "lang": "python", "repo": "tnakamura/SilverTask", "path": "/silvertask-web/kay/sessions/middleware.py", "mode": "spm", "license": "MIT", "source": "the-stack-v2" }
<|fim_prefix|># repo: great-expectations/great_expectations path: /great_expectations/core/usage_statistics/schemas.py "$schema": SCHEMA, "title": "anonymized-test-yaml-config-payload", "definitions": { "anonymized_string": anonymized_string_schema, "anonymized_class_info": anonymized_class_inf...
code_fim
hard
{ "lang": "python", "repo": "great-expectations/great_expectations", "path": "/great_expectations/core/usage_statistics/schemas.py", "mode": "psm", "license": "Apache-2.0", "source": "the-stack-v2" }
<|fim_suffix|>anonymized_expectation_configuration_builder_schema = { "$schema": SCHEMA, "title": "anonymized-expectation-configuration-builder", "definitions": { "anonymized_string": anonymized_string_schema, }, "type": "object", "properties": { "parent_class": {"type": "string"...
code_fim
hard
{ "lang": "python", "repo": "great-expectations/great_expectations", "path": "/great_expectations/core/usage_statistics/schemas.py", "mode": "spm", "license": "Apache-2.0", "source": "the-stack-v2" }
<|fim_prefix|># repo: mpavezb/CarND-Advanced-Lane-Lines path: /main.py from src.logger import Log from src.lane_tracker import LaneLinesTracker from examples import * <|fim_suffix|> input_file = "project_video.mp4" output_file = "output_videos/project_video.mp4" tracker = LaneLinesTracker() clip = tra...
code_fim
medium
{ "lang": "python", "repo": "mpavezb/CarND-Advanced-Lane-Lines", "path": "/main.py", "mode": "psm", "license": "MIT", "source": "the-stack-v2" }
<|fim_suffix|> Log.debug_enabled = False # RunCalibrationExample() # RunDistortionCorrectionExample() # RunEdgeDetectionExample() # RunPerspectiveTransformExample() # RunLaneFittingExample() # RunFullPipelineExample() ProcessProjectVideo(subclip_seconds=None) if __name__ == "__main__":...
code_fim
hard
{ "lang": "python", "repo": "mpavezb/CarND-Advanced-Lane-Lines", "path": "/main.py", "mode": "spm", "license": "MIT", "source": "the-stack-v2" }
<|fim_suffix|>qs = MyModel.objects qs = q.filter(field__regex=r'^(An?|The) +')<|fim_prefix|># repo: andrewp-as-is/django-examples path: /Models/querysets/Field lookups/__regex/tests.py #!/usr/bin/env python from .models import MyModel <|fim_middle|>""" https://docs.djangoproject.com/en/dev/ref/models/querysets/#regex...
code_fim
medium
{ "lang": "python", "repo": "andrewp-as-is/django-examples", "path": "/Models/querysets/Field lookups/__regex/tests.py", "mode": "spm", "license": "Unlicense", "source": "the-stack-v2" }
<|fim_prefix|># repo: andrewp-as-is/django-examples path: /Models/querysets/Field lookups/__regex/tests.py #!/usr/bin/env python from .models import MyModel <|fim_suffix|>qs = MyModel.objects qs = q.filter(field__regex=r'^(An?|The) +')<|fim_middle|>""" https://docs.djangoproject.com/en/dev/ref/models/querysets/#regex...
code_fim
medium
{ "lang": "python", "repo": "andrewp-as-is/django-examples", "path": "/Models/querysets/Field lookups/__regex/tests.py", "mode": "psm", "license": "Unlicense", "source": "the-stack-v2" }
<|fim_prefix|># repo: KKowalewski24/MUM path: /Task4/Program/module/k_means.py from datetime import datetime from typing import List, Tuple import matplotlib.pyplot as plt import numpy as np from sklearn.cluster import KMeans from sklearn.metrics import silhouette_score from module.LatexGenerator import LatexGenerat...
code_fim
hard
{ "lang": "python", "repo": "KKowalewski24/MUM", "path": "/Task4/Program/module/k_means.py", "mode": "psm", "license": "MIT", "source": "the-stack-v2" }
<|fim_suffix|> save_latex: bool = False) -> None: scores_clusters_numbers: List[Tuple[List[float], List[float]]] = [] scores_iter_numbers: List[Tuple[List[float], List[float]]] = [] for cluster_value in CLUSTERS_NUMBER: k_means = KMeans( n_clusters=cluster_value ...
code_fim
hard
{ "lang": "python", "repo": "KKowalewski24/MUM", "path": "/Task4/Program/module/k_means.py", "mode": "spm", "license": "MIT", "source": "the-stack-v2" }
<|fim_prefix|># repo: r3gh/hackathonRioHeatMap path: /server/tweets/src/stream_twitter.py import tweepy from tweepy import OAuthHandler from tweepy.streaming import StreamListener import numpy as np import json import numpy as np from tweets.src.functions import * class StreamTwitterGenerator: def __init__(self): ...
code_fim
medium
{ "lang": "python", "repo": "r3gh/hackathonRioHeatMap", "path": "/server/tweets/src/stream_twitter.py", "mode": "psm", "license": "MIT", "source": "the-stack-v2" }
<|fim_suffix|> consumer_key = tokens['consumer_key'] consumer_secret = tokens['consumer_secret'] access_token = tokens['access_token'] access_secret = tokens['access_secret'] auth = OAuthHandler(consumer_key, consumer_secret) auth.set_access_token(access_token, access_secret) api = ...
code_fim
medium
{ "lang": "python", "repo": "r3gh/hackathonRioHeatMap", "path": "/server/tweets/src/stream_twitter.py", "mode": "spm", "license": "MIT", "source": "the-stack-v2" }
<|fim_suffix|> soup = htmlRequest(url) genreList = soup.select('.breadcrumb > li')[1].contents[1] genreName = genreList.contents[0] return genreName """ # Functions which are 3 functions below from here gives exact URL # however it is little bit slower """ def getCatIDLetters(l,url): i = url.rfind("id") catID =...
code_fim
hard
{ "lang": "python", "repo": "ferhatyaman/appStore", "path": "/appReader.py", "mode": "spm", "license": "MIT", "source": "the-stack-v2" }
<|fim_prefix|># repo: ferhatyaman/appStore path: /appReader.py import urllib.request from bs4 import BeautifulSoup import string, sys, sqlite3, time #if connection lost or server waits the program #try to catch error and sleep until error is gone def htmlRequest(url): success= False while not success: tr...
code_fim
hard
{ "lang": "python", "repo": "ferhatyaman/appStore", "path": "/appReader.py", "mode": "psm", "license": "MIT", "source": "the-stack-v2" }
<|fim_prefix|># repo: 1748276037/Django path: /bookmanager/book/views.py from django.shortcuts import render <|fim_suffix|> context = { 'name':'想了解更多吗?点击我哦' } # 请求 # 参数2: 模板文件 return render(request,'book/index.html',context=context)<|fim_middle|># Create your views here. from django.http i...
code_fim
medium
{ "lang": "python", "repo": "1748276037/Django", "path": "/bookmanager/book/views.py", "mode": "psm", "license": "MIT", "source": "the-stack-v2" }
<|fim_suffix|> context = { 'name':'想了解更多吗?点击我哦' } # 请求 # 参数2: 模板文件 return render(request,'book/index.html',context=context)<|fim_prefix|># repo: 1748276037/Django path: /bookmanager/book/views.py from django.shortcuts import render <|fim_middle|># Create your views here. from django.http i...
code_fim
medium
{ "lang": "python", "repo": "1748276037/Django", "path": "/bookmanager/book/views.py", "mode": "spm", "license": "MIT", "source": "the-stack-v2" }
<|fim_suffix|> # Fourier transform fft2 = staticmethod(npa.fft.fft2) backend = NumpyBackend() def set_backend(name): """ Set the backend for the simulations. This function monkey-patches the backend object by changing its class. This way, all methods of the backend object will be replace...
code_fim
hard
{ "lang": "python", "repo": "SiEPIC/legume", "path": "/legume/backend.py", "mode": "spm", "license": "MIT", "source": "the-stack-v2" }
<|fim_prefix|># repo: SiEPIC/legume path: /legume/backend.py """ Backend for the simulations. Available backends: - numpy [default] - autograd A backend can be set with the 'set_backend' import legume legume.set_backend("autograd") Numpy is still used with some functionalities; if autograd backend is set, ...
code_fim
hard
{ "lang": "python", "repo": "SiEPIC/legume", "path": "/legume/backend.py", "mode": "psm", "license": "MIT", "source": "the-stack-v2" }
<|fim_prefix|># repo: ably/ably-python path: /ably/types/options.py import random import logging from ably.transport.defaults import Defaults from ably.types.authoptions import AuthOptions log = logging.getLogger(__name__) class Options(AuthOptions): def __init__(self, client_id=None, log_level=0, tls=True, re...
code_fim
hard
{ "lang": "python", "repo": "ably/ably-python", "path": "/ably/types/options.py", "mode": "psm", "license": "Apache-2.0", "source": "the-stack-v2" }
<|fim_suffix|> @property def loop(self): return self.__loop # RTC1b @property def auto_connect(self): return self.__auto_connect @property def connection_state_ttl(self): return self.__connection_state_ttl @connection_state_ttl.setter def connection_state_tt...
code_fim
hard
{ "lang": "python", "repo": "ably/ably-python", "path": "/ably/types/options.py", "mode": "spm", "license": "Apache-2.0", "source": "the-stack-v2" }
<|fim_prefix|># repo: tiabas/django-forms path: /form_engine/views.py for_field_type(field_type) if not template_field_form: raise HttpResponseNotFound("<p>Invalid field type: </p>" % request.POST.get('qtype')) form_template = get_object_or_404(Survey, pk=form_id) item_forms = forms_for_survey_no_prefix(request, ...
code_fim
hard
{ "lang": "python", "repo": "tiabas/django-forms", "path": "/form_engine/views.py", "mode": "psm", "license": "MIT", "source": "the-stack-v2" }
End of preview. Expand in Data Studio

QuickCoder-Dataset

This dataset repository stores upload-ready JSONL training checkpoints for code completion and fill-in-the-middle training. Checkpoints are appended in approximately 20 GiB units so they can also be copied to Google Drive and loaded from Colab/H100 training jobs. New checkpoints use one JSONL file per 20 GiB checkpoint. The long-term target is 400 GiB total mirrored to Hugging Face and Google Drive.

Current Upload Status

Only validation-passing checkpoints should be uploaded or used for training.

Checkpoint Size JSONL In-bundle duplicates Status
checkpoint_20260611_104104_bundle01_20g 19.04 GiB valid, single JSONL 0 uploaded to Hugging Face; Google Drive cloud verified by rclone SHA256/size check
checkpoint_20260611_112534_bundle01_20g 19.11 GiB valid, single JSONL 0 uploaded to Hugging Face; Google Drive cloud verified by rclone SHA256/size check
checkpoint_20260611_143422_bundle01_20g 19.19 GiB valid, single JSONL 0 uploaded to Hugging Face; Google Drive upload pending user confirmation

Current Hugging Face dataset repo:

aisamdasu/QuickCoder-Dataset

Repository Layout

README.md
dataset_guide/
tokenizer/
dense/
moe/
dataset/
  checkpoint_YYYYMMDD_HHMMSS_bundleNN_20g/
    dataset/
      checkpoint_YYYYMMDD_HHMMSS_bundleNN_20g.jsonl
dataset_guide/
  checkpoint_reports/
    checkpoint_YYYYMMDD_HHMMSS_bundleNN_20g/
      MANIFEST.json
      BALANCE_REPORT.md
      CHECKSUMS.txt

Record Format

Each line is one UTF-8 JSON object. The main training payload is text. Metadata can appear in top-level fields and in meta.

Common fields:

  • text: canonical training string.
  • domain: task family such as code_fim or code_gen.
  • difficulty: coarse difficulty bucket.
  • meta.lang: programming language.
  • meta.repo, meta.path, meta.license, meta.source: source metadata when available.
  • meta.mode: FIM ordering such as psm or spm.

FIM examples use explicit special tokens:

<|fim_prefix|>{prefix}<|fim_suffix|>{suffix}<|fim_middle|>{middle}

Some FIM shards may use suffix-prefix-middle ordering for robustness.

Source And License Notes

This is a mixed code dataset. Bundle 1 contains generated/unknown continuation records and the-stack-v2 FIM records with per-record source metadata such as repository, path, and license where available. Consumers should filter by meta.license, meta.source, and project policy before redistribution or training.

The dataset card intentionally uses license: other because this repository contains mixed-source records rather than one uniform license.

Validation Contract

A checkpoint is upload-ready only when:

  • JSONL parsing succeeds.
  • Empty text count is zero.
  • In-bundle duplicate count is zero.
  • The checkpoint folder contains only one file: dataset/<checkpoint>.jsonl. Legacy checkpoints may contain several JSONL parts only until they are repackaged.
  • MANIFEST.json, BALANCE_REPORT.md, and CHECKSUMS.txt are present under dataset_guide/checkpoint_reports/<checkpoint>/.
  • Shared tokenizer/, dense/, and moe/ documentation lives at repository root, outside checkpoint folders.

Do not train on a checkpoint marked Not Upload Ready.

Downloads last month
74