[Bug] ModuleNotFoundError: No module named 'datasets' on latest docker build

[manishiitg]

Checklist

• 1. I have searched related issues but cannot get the expected help.
• 2. The bug has not been fixed in the latest version.
• 3. Please note that if the bug-related issue you submitted lacks corresponding environment info and a minimal reproducible demo, it will be challenging for us to reproduce and resolve the issue, reducing the likelihood of receiving feedback.
• 4. If the issue you raised is not a bug but a question, please raise a discussion at https://github.com/sgl-project/sglang/discussions/new/choose Otherwise, it will be closed.
• 5. Please use English, otherwise it will be closed.

Describe the bug

when i usd docker i get this error

inference_engine  | [2025-02-12 12:54:32] INFO:     127.0.0.1:57952 - "POST /v1/chat/completions HTTP/1.1" 200 OK
inference_engine  | [2025-02-12 12:54:32 TP0] Scheduler hit an exception: Traceback (most recent call last):
inference_engine  |   File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 1796, in run_scheduler_process
inference_engine  |     scheduler.event_loop_overlap()
inference_engine  |   File "/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py", line 116, in decorate_context
inference_engine  |     return func(*args, **kwargs)
inference_engine  |   File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 496, in event_loop_overlap
inference_engine  |     batch = self.get_next_batch_to_run()
inference_engine  |   File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 877, in get_next_batch_to_run
inference_engine  |     new_batch = self.get_new_batch_prefill()
inference_engine  |   File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 898, in get_new_batch_prefill
inference_engine  |     self.move_ready_grammar_requests()
inference_engine  |   File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 1565, in move_ready_grammar_requests
inference_engine  |     req.grammar = req.grammar.result(timeout=0.05)
inference_engine  |   File "/usr/lib/python3.10/concurrent/futures/_base.py", line 458, in result
inference_engine  |     return self.__get_result()
inference_engine  |   File "/usr/lib/python3.10/concurrent/futures/_base.py", line 403, in __get_result
inference_engine  |     raise self._exception
inference_engine  |   File "/usr/lib/python3.10/concurrent/futures/thread.py", line 58, in run
inference_engine  |     result = self.fn(*self.args, **self.kwargs)
inference_engine  |   File "/sgl-workspace/sglang/python/sglang/srt/constrained/base_grammar_backend.py", line 53, in init_value
inference_engine  |     entry.value = self.init_value_impl(key)
inference_engine  |   File "/sgl-workspace/sglang/python/sglang/srt/constrained/outlines_backend.py", line 173, in init_value_impl
inference_engine  |     guide = RegexGuide.from_regex(regex, self.outlines_tokenizer)
inference_engine  |   File "/usr/local/lib/python3.10/dist-packages/outlines/fsm/guide.py", line 92, in from_regex
inference_engine  |     return super().from_regex(
inference_engine  |   File "/usr/local/lib/python3.10/dist-packages/outlines_core/fsm/guide.py", line 212, in from_regex
inference_engine  |     ) = _create_states_mapping(
inference_engine  |   File "/usr/local/lib/python3.10/dist-packages/outlines/fsm/guide.py", line 76, in cached_create_states_mapping
inference_engine  |     return uncached_create_states_mapping(regex_string, tokenizer, *args, **kwargs)
inference_engine  |   File "/usr/local/lib/python3.10/dist-packages/outlines_core/fsm/guide.py", line 141, in create_states_mapping
inference_engine  |     return create_states_mapping_from_fsm(regex_fsm, tokenizer, frozen_tokens)
inference_engine  |   File "/usr/local/lib/python3.10/dist-packages/outlines_core/fsm/guide.py", line 178, in create_states_mapping_from_fsm
inference_engine  |     states_to_token_maps, empty_token_ids = create_fsm_index_tokenizer(
inference_engine  |   File "/usr/local/lib/python3.10/dist-packages/outlines_core/fsm/regex.py", line 473, in create_fsm_index_tokenizer
inference_engine  |     tokens_to_token_ids, empty_token_ids = reduced_vocabulary(tokenizer)
inference_engine  |   File "/usr/local/lib/python3.10/dist-packages/outlines/models/transformers.py", line 117, in __hash__
inference_engine  |     from datasets.fingerprint import Hasher
inference_engine  | ModuleNotFoundError: No module named 'datasets'
inference_engine  | 

Reproduction

using docker

Environment

h100

[jhinpan]

You can install the missing module datasets on your own. We will try to include that in the next version of docker build. cc @zhyncs