label stringlengths 14 52 | messages listlengths 1 1 | metadata dict | prompt listlengths 1 1 |
|---|---|---|---|
theacodes__nox-310 | [
{
"content": "Session.run finds incorrect Python \n**Describe the bug**\r\n`session.run('python', ..., ....)` can not find the *virtual* environments python. \r\n\r\n**How to reproduce**\r\nHere is my noxfile.py. The last line is the where the issue occurs. There are **no** issues in creating a conda environme... | {
"base_commit": "8dcefbf8f59293bc1925dd13376dd6d3aa66e53a",
"benchmark": "SWE-ReBench-V2",
"image_name": "docker.io/swerebenchv2/theacodes-nox:310-8dcefbf",
"instance_id": "theacodes__nox-310",
"privileged_info": {
"pi_artifact_sha256": "a67a0b75dcbc6bc8b48f5d352dba7b78138b94379a148c9ec187ec3ea326cc96",
... | [
{
"content": "Session.run finds incorrect Python \n**Describe the bug**\r\n`session.run('python', ..., ....)` can not find the *virtual* environments python. \r\n\r\n**How to reproduce**\r\nHere is my noxfile.py. The last line is the where the issue occurs. There are **no** issues in creating a conda environme... |
berriai__litellm-1200 | [
{
"content": "[Feature]: acomplete see explicit args instead of *args, **kwargs on acomplete,\n### The Feature\n\n> would love to see explicit args instead of *args, **kwargs on acomplete, like you have in complete\n\n### Motivation, pitch\n\neveryone loves intellisense \n\n### Twitter / LinkedIn details\n\n_N... | {
"base_commit": "4cfa010dbda2cffa7a403d6e13a6156904bd056c",
"benchmark": "SWE-ReBench-V2",
"image_name": "docker.io/swerebenchv2/berriai-litellm:1200-4cfa010",
"instance_id": "berriai__litellm-1200",
"privileged_info": {
"pi_artifact_sha256": "6467347961fd5f159af5ac14c3ca8020365315d1751d8f033baebf6c09d2b... | [
{
"content": "[Feature]: acomplete see explicit args instead of *args, **kwargs on acomplete,\n### The Feature\n\n> would love to see explicit args instead of *args, **kwargs on acomplete, like you have in complete\n\n### Motivation, pitch\n\neveryone loves intellisense \n\n### Twitter / LinkedIn details\n\n_N... |
pybamm-team__pybamm-856 | [
{
"content": "v0.2.0 release\nFinal things to do for release:\r\n\r\n- update version number\r\n- update version number in docs\r\n- update authors in docs\r\n- update changelog section title from \"unreleased\"\r\n\r\nanything else? @tlestang is all the pip stuff automatic?",
"role": "user"
}
] | {
"base_commit": "4bf599043c9bd070333e224180abc455c8127940",
"benchmark": "SWE-ReBench-V2",
"image_name": "docker.io/swerebenchv2/pybamm-team-pybamm:856-4bf5990",
"instance_id": "pybamm-team__pybamm-856",
"privileged_info": {
"pi_artifact_sha256": "eccabe245613d884f593a91117fa57cfacbe73772c89e6e945eec0f73... | [
{
"content": "v0.2.0 release\nFinal things to do for release:\r\n\r\n- update version number\r\n- update version number in docs\r\n- update authors in docs\r\n- update changelog section title from \"unreleased\"\r\n\r\nanything else? @tlestang is all the pip stuff automatic?",
"role": "user"
}
] |
agnostiqhq__covalent-1251 | [
{
"content": "Corrupted config file\n### Environment\n\n- **Covalent version**: 0.198.0\r\n- **Python version**: 3.8.13\r\n- **Operating system**: MacOS 12.3.1\r\n\n\n### What is happening?\n\nIt seems the corrupted config file issue has crept up again. This is most likely due to performing multiple writes from... | {
"base_commit": "a7701b756f458086a7f70a4aa7eb263aed07a9d4",
"benchmark": "SWE-ReBench-V2",
"image_name": "docker.io/swerebenchv2/agnostiqhq-covalent:1251-a7701b7",
"instance_id": "agnostiqhq__covalent-1251",
"privileged_info": {
"pi_artifact_sha256": "d5f1a70b33f85f6e1ec96cd0d1a29b4ee77422501c23186e978e7... | [
{
"content": "Corrupted config file\n### Environment\n\n- **Covalent version**: 0.198.0\r\n- **Python version**: 3.8.13\r\n- **Operating system**: MacOS 12.3.1\r\n\n\n### What is happening?\n\nIt seems the corrupted config file issue has crept up again. This is most likely due to performing multiple writes from... |
pyomo__pyomo-2263 | [
{
"content": "gdp.fix_disjuncts assumes all Disjuncts are set and doesn't die very gracefully if they aren't\n## Summary\r\n\r\nThis is issue is mainly a question, which is: Should the `gdp.fix_disjuncts` transformation be a transformation from a GDP to a MI(N)LP? Historically, this seems to have been assumed: ... | {
"base_commit": "7b76b1dc4483a953bcd3019de5700416a7533da4",
"benchmark": "SWE-ReBench-V2",
"image_name": "docker.io/swerebenchv2/pyomo-pyomo:2263-7b76b1d",
"instance_id": "pyomo__pyomo-2263",
"privileged_info": {
"pi_artifact_sha256": "202f068d9e681747fa9303e378ec06e1f61e20bb23cb095dd302cf1f27c54cc8",
... | [
{
"content": "gdp.fix_disjuncts assumes all Disjuncts are set and doesn't die very gracefully if they aren't\n## Summary\r\n\r\nThis is issue is mainly a question, which is: Should the `gdp.fix_disjuncts` transformation be a transformation from a GDP to a MI(N)LP? Historically, this seems to have been assumed: ... |
cs-si__eodag-1717 | [
{
"content": "Two consecutive executions will fully download twice the same products\n**Describe the bug**\nWhen running an eodag request twice in a row, products are being fully downloaded twice while eodag would be expected to detect that files already exist locally and jump on to next search result.\n\n**Cod... | {
"base_commit": "036b86bbefeed905c9962a7a4bf7bca8258246fb",
"benchmark": "SWE-ReBench-V2",
"image_name": "docker.io/swerebenchv2/cs-si-eodag:1717-036b86b",
"instance_id": "cs-si__eodag-1717",
"privileged_info": {
"pi_artifact_sha256": "47ab74bd56cc5a605fde1d6d7d2d2bd394309e739b96a771143d973d49135486",
... | [
{
"content": "Two consecutive executions will fully download twice the same products\n**Describe the bug**\nWhen running an eodag request twice in a row, products are being fully downloaded twice while eodag would be expected to detect that files already exist locally and jump on to next search result.\n\n**Cod... |
aws-cloudformation__cfn-lint-3633 | [
{
"content": "Is this valid Cloudformation and a cfn-lint bug?\n**cfn-lint Version **\r\n\r\n0.87.3\r\n\r\nSyntax:\r\n\r\n// add a pseudo element to array (/null) so we can rely on intrinsic length to pop last element we want\r\n `Id: !Select [ !Length !Split [\"/\", !Sub \"${AWS::StackName}-v1\"], !Split [ \"/... | {
"base_commit": "754f13207650a93bab351d0c31bfd5324d3cf8e4",
"benchmark": "SWE-ReBench-V2",
"image_name": "docker.io/swerebenchv2/aws-cloudformation-cfn-lint:3633-754f132",
"instance_id": "aws-cloudformation__cfn-lint-3633",
"privileged_info": {
"pi_artifact_sha256": "6eb5dd18e0096a60bb88595b01c832a3ac8cb... | [
{
"content": "Is this valid Cloudformation and a cfn-lint bug?\n**cfn-lint Version **\r\n\r\n0.87.3\r\n\r\nSyntax:\r\n\r\n// add a pseudo element to array (/null) so we can rely on intrinsic length to pop last element we want\r\n `Id: !Select [ !Length !Split [\"/\", !Sub \"${AWS::StackName}-v1\"], !Split [ \"/... |
amaranth-lang__amaranth-1115 | [
{
"content": "Deprecate and remove `ValueCastable.lowermethod`\nThis isn't actually serving any useful purpose.\r\n\r\n- Deprecation PR: N/A\r\n- Removal PR: N/A (for milestone 0.6)",
"role": "user"
}
] | {
"base_commit": "0056e982c5a85b26935b8c750cc00cf4b1e66ebc",
"benchmark": "SWE-ReBench-V2",
"image_name": "docker.io/swerebenchv2/amaranth-lang-amaranth:1115-0056e98",
"instance_id": "amaranth-lang__amaranth-1115",
"privileged_info": {
"pi_artifact_sha256": "62f46881705000dff564b0340882be680257dcf6b439ab1... | [
{
"content": "Deprecate and remove `ValueCastable.lowermethod`\nThis isn't actually serving any useful purpose.\r\n\r\n- Deprecation PR: N/A\r\n- Removal PR: N/A (for milestone 0.6)",
"role": "user"
}
] |
xknx__xknx-1466 | [
{
"content": "Feature Request : Integration of DPT 235.001 / DPT_Tariff_ActiveEnergy\nHello, do you think it's possible to integrate the DPT 235.001. Is a 6bytes DPT. This contains DPT 13.010 and 5.006+1 data validity byte. \r\n\r\n![image](https://user-images.githubusercontent.com/10090794/208304366-843c6c51-d... | {
"base_commit": "ed433308d43dd2b48f145533deb4267026b0ed13",
"benchmark": "SWE-ReBench-V2",
"image_name": "docker.io/swerebenchv2/xknx-xknx:1466-ed43330",
"instance_id": "xknx__xknx-1466",
"privileged_info": {
"pi_artifact_sha256": "987a5210cacb3a20c1a43b2de4d10422af6a2f78c2fbb89e3687f3ddd371abf0",
"p... | [
{
"content": "Feature Request : Integration of DPT 235.001 / DPT_Tariff_ActiveEnergy\nHello, do you think it's possible to integrate the DPT 235.001. Is a 6bytes DPT. This contains DPT 13.010 and 5.006+1 data validity byte. \r\n\r\n![image](https://user-images.githubusercontent.com/10090794/208304366-843c6c51-d... |
textualize__rich-1804 | [{"content":"[BUG] repr fails for pd.Series in Jupyter with pretty.install()\n**Describe the bug**\r(...TRUNCATED) | {"base_commit":"d4450d9e43375e43614ef81eaaa5b9c18a8f278a","benchmark":"SWE-ReBench-V2","image_name":(...TRUNCATED) | [{"content":"[BUG] repr fails for pd.Series in Jupyter with pretty.install()\n**Describe the bug**\r(...TRUNCATED) |
End of preview. Expand in Data Studio
OPSD PI MultiRepo — Student-Fail
本仓库公开发布之前已生成的多仓库 PI，供 OPSD 训练下载。1024 条任务、122 个仓库，每题包含 5 个阶段 × 3 种强度，共 15 条 PI 内容。
数据身份
从历史 student 的确认 unresolved 任务中选择，1024 条全部为确认未解出；不包含 infrastructure-error 筛选结果。
- SWE-ReBench-V2：824 条。
- SWE-Gym：200 条。
- 抽样 seed：42。原始选择日期：2026-09-22；新增 ReBench PI 合成批次：2026-09-24；本次公开发布日期：2026-10-15。
- 历史筛选 student：
jiaxingx/privilege-code-opsd-ckpts/qwen35-9b-expert-sft-131k-lora64-block28-2996448/checkpoint-best，agent：mini-swe-agent 2.3.0。这个标识记录历史模型身份，不保证该模型当前可公开下载。 - 另一组：LSW142857/OPSD-PI-MultiRepo-All-Outcomes-1024。
两组是替代实验组，不是 train/test 划分。它们重叠 460 个 instance ID、466 个相同 problem statement；各组内部没有重复 ID 或问题文本。新 SFT 模型的成功/失败结果可能不同，不能把历史 student-fail 当作新模型已重新评测的结论。
文件与下载
data/train.jsonl：完整训练数据，顶层字段label/messages/metadata/prompt。data/manifest.json：来源、抽样信息、PI 校验警告、文件哈希、逐题 PI 来源标识。data/repo_counts.json：122 个仓库的逐仓库条数。SHA256SUMS：发布文件校验值。
hf download LSW142857/OPSD-PI-MultiRepo-Student-Fail-1024 --repo-type dataset --local-dir ./OPSD-PI-MultiRepo-Student-Fail-1024
cd OPSD-PI-MultiRepo-Student-Fail-1024
sha256sum -c SHA256SUMS
wc -l data/train.jsonl
为严格复现，请将 --revision <commit SHA> 加入下载命令，固定仓库版本。
JSONL 可直接逐行读取，或使用 datasets.load_dataset("json", data_files="data/train.jsonl", split="train")。OPSD 使用 prompt 作为公开任务输入，并按配置从下列路径选择 teacher PI：
metadata.privileged_info.pi_options
EXPLORE / REPRODUCE / DIAGNOSE / EDIT / VERIFY
WEAK / MEDIUM / STRONG
content / evidence_ids / name
metadata 另含 instance_id、benchmark、repo、base_commit、repository_path、image_name、problem_statement。环境字段是平铺格式。此仓库不包含模型权重、容器镜像、student 轨迹、PI 合成完整轨迹或原始 gold/test patch 字段。
PI 来源和质量边界
SWE-ReBench-V2 PI 来自历史 deepseek/deepseek-flash 合成批次中已经 accepted 的输出；SWE-Gym 复用已有 schema 1.1 PI。合成目录名称中虽有 medium，本发布包含 WEAK、MEDIUM、STRONG 三档，未丢弃其他阶段或档位。
本版本 data/train.jsonl 与 2026-09-30 clip122 冻结训练文件逐字节一致。公开 manifest 已去除旧机器私有路径，因此其 SHA 与旧 manifest 不同；原 manifest SHA 保存在 original_local_training_manifest_sha256。
训练 JSONL SHA-256：80130f0317a828960a2db881fa674f1370c3e41f7e86823a40488b5071b6bc96。
保留历史合成批次的接受策略和校验警告：本组有 824 条任务记录了警告，共 13071 条警告；详情在 manifest 的 provenance[].validation_warnings。这些包括阶段纯度、与测试代码的短片段重合及证据审计警告。数据通过结构和完整性检查不等于全部 PI 语义都正确。PI 含有解题特权信息，只应按实验设计供 teacher 使用，不应直接拼入无 PI 的 student 或评测输入。
来源与使用条款
任务来自 SWE-Gym 和 SWE-ReBench-V2 及其各自上游代码仓库。每条记录保留 benchmark/repo/base_commit 用于追溯。此发布不为上游内容重新授予统一许可证；使用和再分发时应分别遵守相应原始数据、代码与生成服务条款。
只发布完成训练所需的任务和 PI 及可移植来源说明；旧机器路径不构成下载依赖。manifest 中 PI artifact hash 指向历史生成文件，相关原始生成文件没有另行打包进本仓库。
- Downloads last month
- 252