Datasets:

jonhlin
/

stg-rl-train-envs

Error code:   StreamingRowsError
Exception:    ValueError
Message:      Expected object or value
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 290, in _generate_tables
                  pa_table = paj.read_json(
                      io.BytesIO(batch), read_options=paj.ReadOptions(block_size=block_size)
                  )
                File "pyarrow/_json.pyx", line 342, in pyarrow._json.read_json
                File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
                File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
                  raise convert_status(status)
              pyarrow.lib.ArrowInvalid: JSON parse error: Column(/tools/[]/preconditions/[]/allowed_values/[]) changed from number to string in row 0
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
                  return get_rows(
                      dataset=dataset,
                  ...<4 lines>...
                      column_names=column_names,
                  )
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
                  rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
                File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
                  yield from ds.decode(False) if ds.features else ds
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2840, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 304, in _generate_tables
                  batch = json_encode_fields_in_json_lines(original_batch, json_field_paths)
                File "/usr/local/lib/python3.14/site-packages/datasets/utils/json.py", line 111, in json_encode_fields_in_json_lines
                  examples = [ujson_loads(line) for line in original_batch.splitlines()]
                              ~~~~~~~~~~~^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/utils/json.py", line 20, in ujson_loads
                  return pd.io.json.ujson_loads(*args, **kwargs)
                         ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
              ValueError: Expected object or value

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

YAML Metadata Warning:The task_categories "agent" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other

STG RL 训练环境精选集(12 环境 / 240 任务)

由 STG(Synthetic Task Generation)管线合成的中文工具调用环境,专为 agentic RL(verl GRPO + step-level reward) 训练筛选的高质量子集。每个环境 = 一个自包含的 SQLite 业务库(schema + 样本数据)+ 工具定义 + 程序化验收任务。

规模

环境:12 个(envs/,每个含 EnvironmentDef JSON + 可执行 SQLite 库)
任务:240 个(tasks/,每个含中文描述、ground-truth solution、声明式验收 check)
- 难度分布:36 easy / 84 medium / 84 hard / 36 extreme
- 类型分布:60 create / 60 modify / 60 query / 60 refusal
业务域:8 个——airline, ota, banking, retail, delivery, instore, gaming, pet_care
seed:12 个(seeds/,合成输入 YAML,含业务域描述、软约束维度、任务类型与难度偏好)

筛选标准

从 34 个合成环境中按 6 维 rubric 静态评估选出(详见 docs/rl_env_selection.md 口径):

reward 可判定且解唯一(core check 为确定值、初始态为假、无多解钉死/自拟文本假阴性)
关键信息在库不在描述,必须靠工具检索
软约束/陷阱密度("表面完成但违反约束"的失败模式)
数据自洽且有干扰项
难度落在学习区(medium/hard/extreme 的 create/modify 段)
非同构、领域多样

结构

envs/
  env_seed_<domain>_<NNN>.json   # EnvironmentDef:schema/样本数据/工具定义
  env_seed_<domain>_<NNN>.db     # 建好库的 SQLite(可直接执行工具/验收)
tasks/
  env_seed_<domain>_<NNN>_NN.json  # TaskDef(见下),env_id 指向 envs/
seeds/
  seed_<domain>_<NNN>.yaml       # 合成输入(可复用 STG 管线继续合成)

任务格式(TaskDef)

description:中文自然语言用户请求(不含工具名/表名/编号)
task_type:create | modify | query | refusal(拒绝类 core/bonus/solution 为空)
difficulty:easy | medium | hard | extreme
acceptance:core(必须达成)+ bonus(软约束加分)声明式 check, 五种 kind:row_state / row_exists / row_absent / aggregate / predicate
solution:正确完成全部子目标的工具调用序列(沙箱重放验证过)

使用说明

RL 训练(本数据集主用途):配合 step-level reward(solution 链提供 oracle 图距离信号);RL v1 口径 query/refusal 任务不打包,每个环境的有效训练题量为 10 题(5 create + 5 modify)。
policy.txt:环境目录中未附 policy 文本,训练前需用 STG 管线 stg policy-gen 补生成(进入 system prompt)。
验收判定:可用 STG 的 sandbox + check_evaluator 在 .db 上执行。

已知问题(不影响主用途)

env_seed_ota_001 的 _16 夜数存在歧义(2 晚/3 晚),建议补一句描述;
env_seed_retail_002 的 _17 core 含无出处的措辞检查(无需补差价);
env_seed_delivery_004 的 _04 选店存在多解(描述未要求"最便宜",check 写死)。

Downloads last month: 60

Total file size:

1.17 MB