[None][feat] Add gpt-oss GSM8K test. #6732
Conversation
Signed-off-by: Tracin <10434017+Tracin@users.noreply.github.com>
📝 WalkthroughWalkthroughAdded a new boolean flag Changes
Sequence Diagram(s)sequenceDiagram
participant TestRunner
participant TestGPTOSS
participant AccuracyTask
participant LmEvalEvaluator
participant lm_eval
TestRunner->>TestGPTOSS: run test method (e.g., test_w4_1gpu)
TestGPTOSS->>AccuracyTask: update_task_kwargs(task) (sets fewshot_as_multiturn, scores_filter, ...)
TestGPTOSS->>AccuracyTask: task.evaluate(llm, **task_kwargs)
AccuracyTask->>LmEvalEvaluator: evaluate(..., scores_filter=..., fewshot_as_multiturn=...)
LmEvalEvaluator->>lm_eval: lm_eval.evaluate(..., fewshot_as_multiturn=...)
lm_eval-->>LmEvalEvaluator: metrics
LmEvalEvaluator-->>AccuracyTask: aggregated/filtered score
AccuracyTask-->>TestGPTOSS: result/assertions
Estimated code review effort🎯 2 (Simple) | ⏱️ ~8 minutes Possibly related PRs
Suggested labels
Suggested reviewers
Note 🔌 MCP (Model Context Protocol) integration is now available in Early Access!Pro users can now connect to remote MCP servers under the Integrations page to get reviews and chat conversations that understand additional development context. 📜 Recent review detailsConfiguration used: .coderabbit.yaml 📒 Files selected for processing (1)
🚧 Files skipped from review as they are similar to previous changes (1)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
✨ Finishing Touches🧪 Generate unit tests
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
Documentation and Community
|
|
/bot run --disable-fail-fast --add-multi-gpu-test |
|
PR_Github #14563 [ run ] triggered by Bot |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
tests/integration/test_lists/test-db/l0_dgx_h100.yml (1)
123-144: Good addition but TRTLLM backend coverage is missingNice to see a dedicated
gpt_ossauto-trigger block for 4-GPU H100 nodes.
However, unlike the B200 list, TRTLLM back-ends are not included for thetp4/ep4/dp4cases. For consistency (and to catch TRTLLM-specific regressions) consider adding them:+ - accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_w4_4gpus[tp4-TRTLLM] + - accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_w4_4gpus[ep4-TRTLLM] + - accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_w4_4gpus[dp4-TRTLLM]If TRTLLM is intentionally excluded on H100, please document the rationale to avoid confusion later.
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (8)
tensorrt_llm/evaluate/interface.py(1 hunks)tensorrt_llm/evaluate/lm_eval.py(7 hunks)tests/integration/defs/accuracy/accuracy_core.py(2 hunks)tests/integration/defs/accuracy/references/gsm8k.yaml(1 hunks)tests/integration/defs/accuracy/test_llm_api_pytorch.py(7 hunks)tests/integration/test_lists/qa/llm_function_full.txt(1 hunks)tests/integration/test_lists/test-db/l0_b200.yml(1 hunks)tests/integration/test_lists/test-db/l0_dgx_h100.yml(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: Python code should conform to Python 3.8+.
Indent Python code with 4 spaces. Do not use tabs.
Always maintain the namespace when importing in Python, even if only one class or function from a module is used.
Python filenames should use snake_case (e.g., some_file.py).
Python classes should use PascalCase (e.g., class SomeClass).
Python functions and methods should use snake_case (e.g., def my_awesome_function():).
Python local variables should use snake_case. Prefix k for variable names that start with a number (e.g., k_99th_percentile).
Python global variables should use upper snake_case and prefix G (e.g., G_MY_GLOBAL).
Python constants should use upper snake_case (e.g., MY_CONSTANT).
Avoid shadowing variables declared in an outer scope in Python.
Initialize all externally visible members of a Python class in the constructor.
For interfaces that may be used outside a Python file, prefer docstrings over comments.
Comments in Python should be reserved for code within a function, or interfaces that are local to a file.
Use Google style docstrings for Python classes and functions, which can be parsed by Sphinx.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the class docstring.
Avoid using reflection in Python when functionality can be easily achieved without it.
When using try-except blocks in Python, limit the except to the smallest set of errors possible.
When using try-except blocks to handle multiple possible variable types in Python, keep the body of the try as small as possible, using the else block to implement the logic.
Files:
tensorrt_llm/evaluate/interface.pytests/integration/defs/accuracy/test_llm_api_pytorch.pytensorrt_llm/evaluate/lm_eval.pytests/integration/defs/accuracy/accuracy_core.py
**/*.{cpp,h,hpp,cc,cxx,cu,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the current year. This includes .cpp, .h, .cu, .py, and any other source files which are compiled or interpreted.
Files:
tensorrt_llm/evaluate/interface.pytests/integration/defs/accuracy/test_llm_api_pytorch.pytensorrt_llm/evaluate/lm_eval.pytests/integration/defs/accuracy/accuracy_core.py
🧠 Learnings (4)
📓 Common learnings
Learnt from: moraxu
PR: NVIDIA/TensorRT-LLM#6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
📚 Learning: 2025-07-28T17:06:08.621Z
Learnt from: moraxu
PR: NVIDIA/TensorRT-LLM#6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
Applied to files:
tests/integration/test_lists/test-db/l0_b200.ymltests/integration/test_lists/qa/llm_function_full.txttests/integration/test_lists/test-db/l0_dgx_h100.ymltests/integration/defs/accuracy/test_llm_api_pytorch.pytensorrt_llm/evaluate/lm_eval.py
📚 Learning: 2025-08-06T13:58:07.506Z
Learnt from: galagam
PR: NVIDIA/TensorRT-LLM#6487
File: tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py:1-12
Timestamp: 2025-08-06T13:58:07.506Z
Learning: In TensorRT-LLM, test files (files under tests/ directories) do not require NVIDIA copyright headers, unlike production source code files. Test files typically start directly with imports, docstrings, or code.
Applied to files:
tests/integration/test_lists/test-db/l0_b200.ymltests/integration/test_lists/qa/llm_function_full.txttests/integration/defs/accuracy/test_llm_api_pytorch.py
📚 Learning: 2025-07-17T09:01:27.402Z
Learnt from: amitz-nv
PR: NVIDIA/TensorRT-LLM#5616
File: tensorrt_llm/executor/worker.py:375-384
Timestamp: 2025-07-17T09:01:27.402Z
Learning: In tensorrt_llm/executor/worker.py, the LoRA adapter cache optimization logic that checks `is_adapter_in_cpu_cache()` and conditionally passes None for weights/config has a known race condition issue that cannot be solved with simple error handling or verification checks. This is a known limitation that requires a more comprehensive solution.
Applied to files:
tests/integration/test_lists/test-db/l0_b200.yml
🧬 Code Graph Analysis (1)
tensorrt_llm/evaluate/interface.py (1)
tensorrt_llm/evaluate/lm_eval.py (1)
apply_chat_template(54-64)
🔇 Additional comments (22)
tensorrt_llm/evaluate/interface.py (1)
36-42: LGTM! Clean implementation of the new evaluation mode parameter.The addition of
fewshot_as_multiturnparameter follows Python conventions with proper type annotation, reasonable default value, and correct instance attribute assignment.tests/integration/defs/accuracy/references/gsm8k.yaml (1)
162-168: LGTM! Accuracy improvements with new evaluation mode.The consistent improvement from 88.5 to 90.3 across GPT-OSS model variants reflects the benefits of the new
fewshot_as_multiturnevaluation approach for GSM8K tasks.tests/integration/test_lists/qa/llm_function_full.txt (1)
522-534: LGTM! Comprehensive test coverage for GPT-OSS GSM8K evaluation.The new test entries provide thorough coverage across different GPU configurations (1 and 4 GPUs), parallelism strategies (tp4, ep4, dp4), backends (CUTLASS, TRTLLM, TRITON), and quantization schemes (w4, w4a16). This aligns well with the PR objective to add GSM8K tests for the GPT-OSS model.
tests/integration/defs/accuracy/accuracy_core.py (2)
195-199: LGTM! Clean implementation of scores filtering support.The conditional assignment of
scores_filterand its integration into the evaluator call is well-implemented, supporting the new selective metric reporting capability.
301-301: LGTM! Proper class attribute definition for scores filtering.The explicit definition of
SCORES_FILTER = Noneestablishes clear default behavior for GSM8K tasks and follows the established class attribute naming conventions.tests/integration/defs/accuracy/test_llm_api_pytorch.py (8)
2464-2464: LGTM: Clean refactor to class constantThe MODEL_PATH constant is a good improvement over the previous method call approach, making the model path more accessible and efficient.
2466-2471: LGTM: Correct implementation of new evaluation featuresThis method properly implements the core PR objective by enabling the
fewshot_as_multiturnflag and related evaluation parameters. The configuration is appropriate for GSM8K mathematical reasoning tasks with flexible score extraction.
2486-2486: LGTM: Consistent usage of MODEL_PATH constantThe usage of
self.MODEL_PATHis correct and consistent with the refactoring approach.
2497-2497: LGTM: Proper application of task configurationThe call to
self.update_task_kwargs(task)correctly applies the new evaluation parameters before running the GSM8K evaluation.
2535-2535: LGTM: Consistent task configurationProper application of the update_task_kwargs method to configure the GSM8K task with the new evaluation parameters.
2565-2565: LGTM: Complete test coverageThe task configuration is properly applied in the w4a16 test method, ensuring all test variants use the new fewshot_as_multiturn functionality.
2523-2523: LGTM: Consistent MODEL_PATH usageProper usage of the MODEL_PATH constant in the multi-GPU test method.
2554-2554: LGTM: Complete refactoringThe MODEL_PATH constant is correctly used in the w4a16 test method, completing the refactoring.
tensorrt_llm/evaluate/lm_eval.py (9)
135-135: LGTM! Parameter addition follows the existing pattern.The
fewshot_as_multiturnparameter is correctly added with a default value ofFalse, maintaining backward compatibility while enabling the new functionality.
147-147: LGTM! Parameter properly propagated to superclass.The
fewshot_as_multiturnparameter is correctly passed to the parentEvaluatorclass constructor, maintaining the inheritance chain properly.
195-196: LGTM! Method signature enhancement with proper defaults.The addition of
streamingandscores_filterparameters to theevaluatemethod follows good practices:
- Both parameters have sensible defaults (
FalseandNone)- The
scores_filterparameter enables selective metric reporting as described in the AI summary
198-204: LGTM! Proper integration with lm_eval library.The
lm_eval.evaluatecall is updated correctly:
- All new parameters are properly passed through
- The
fewshot_as_multiturnparameter enables the new evaluation mode- Parameter names match the lm_eval API expectations
213-223: LGTM! Flexible scoring logic with proper fallback.The conditional logic for score filtering is well-implemented:
- When
scores_filteris provided, it returns the specific filtered metric- Falls back to the original average accuracy calculation when no filter is specified
- Both paths include appropriate logging for transparency
233-234: LGTM! CLI parameter handling follows the established pattern.The
fewshot_as_multiturnparameter is correctly extracted from kwargs with the appropriate default value, consistent with other CLI parameters in the method.
268-271: LGTM! CLI option definition follows Click conventions.The CLI option for
fewshot_as_multiturnis properly defined:
- Uses
is_flag=Truefor boolean behavior- Has appropriate default value (
False)- Includes descriptive help text
287-290: LGTM! Important validation logic for parameter dependencies.The assertion correctly enforces the business rule that
apply_chat_templatemust beTruewhenfewshot_as_multiturnisTrue. This prevents invalid configurations and provides clear error messaging.
144-144: No change needed: the explicitimport lm_eval.tasksis requiredThe submodule
lm_eval.tasksmust be explicitly imported to populatelm_eval.tasksinsys.modules; otherwise,lm_eval.tasks.TaskManagerandlm_eval.tasks.get_task_dictwould raise an AttributeError.
- tensorrt_llm/evaluate/lm_eval.py line 144: keep
import lm_eval.tasksas is.Likely an incorrect or invalid review comment.
Signed-off-by: Tracin <10434017+Tracin@users.noreply.github.com>
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (1)
tests/integration/test_lists/test-db/l0_dgx_b200.yml (1)
73-81: Optional: add a pipeline-parallel variant if supported (pp4 or tp2pp2).Other tests in this file exercise pp4/tp2pp2 on 4 GPUs. If TestGPTOSS supports it, consider adding one to catch PP-specific regressions.
Example additions (only if supported by TestGPTOSS):
+ - accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_w4_4gpus[pp4-TRTLLM] + - accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_w4_4gpus[tp2pp2-TRTLLM]
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
tests/integration/defs/accuracy/accuracy_core.py(2 hunks)tests/integration/defs/accuracy/test_llm_api_pytorch.py(7 hunks)tests/integration/test_lists/test-db/l0_b200.yml(1 hunks)tests/integration/test_lists/test-db/l0_dgx_b200.yml(1 hunks)
🚧 Files skipped from review as they are similar to previous changes (3)
- tests/integration/test_lists/test-db/l0_b200.yml
- tests/integration/defs/accuracy/accuracy_core.py
- tests/integration/defs/accuracy/test_llm_api_pytorch.py
🧰 Additional context used
🧠 Learnings (2)
📚 Learning: 2025-07-28T17:06:08.621Z
Learnt from: moraxu
PR: NVIDIA/TensorRT-LLM#6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
Applied to files:
tests/integration/test_lists/test-db/l0_dgx_b200.yml
📚 Learning: 2025-08-06T13:58:07.506Z
Learnt from: galagam
PR: NVIDIA/TensorRT-LLM#6487
File: tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py:1-12
Timestamp: 2025-08-06T13:58:07.506Z
Learning: In TensorRT-LLM, test files (files under tests/ directories) do not require NVIDIA copyright headers, unlike production source code files. Test files typically start directly with imports, docstrings, or code.
Applied to files:
tests/integration/test_lists/test-db/l0_dgx_b200.yml
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Pre-commit Check
🔇 Additional comments (4)
tests/integration/test_lists/test-db/l0_dgx_b200.yml (4)
73-81: Solid coverage expansion for GPT-OSS on 4xB200.The tp/ep/dp × CUTLASS/TRTLLM/TRITON matrix looks consistent with the suite’s 4-GPU constraint and post-merge scope. Good placement and naming.
82-82: Confirmed: test_w4a16 only supports “dp4”
TheTestGPTOSS::test_w4a16function is parametrized with a single(4,1,4,True,True,True)entry and theids=["dp4"]marker, so listing only thedp4variant in the YAML is intentional and correct.
73-81: TRITON MOE backend gating is already in place
The “TRITON” option here refers to the fused-MOE Triton-kernels backend (not Triton Inference Server), and each test skips appropriately when it’s unavailable or the SM version is too old:• tests/integration/defs/accuracy/test_llm_api_pytorch.py
– test_w4a8_mxfp4 (≈ lines 2231–2236): skips ifIS_TRITON_KERNELS_AVAILABLEis false or SM < 90
– test_w4_1gpu (≈ lines 2479–2481): skips ifIS_TRITON_KERNELS_AVAILABLEis false
– test_w4_4gpus (≈ lines 2511–2516): skips ifIS_TRITON_KERNELS_AVAILABLEis false ortp_size != ep_sizeNo Triton Inference Server setup or additional skip markers are needed here.
73-81: Test W4_4gpus parameter IDs are defined correctlyVerified in tests/integration/defs/accuracy/test_llm_api_pytorch.py that TestGPTOSS::test_w4_4gpus is decorated with:
@pytest.mark.parametrize("moe_backend", ["CUTLASS","TRTLLM","TRITON"])@pytest.mark.parametrize("tp_size,pp_size,ep_size,attention_dp,cuda_graph,overlap_scheduler", …, ids=["tp4","ep4","dp4"])This produces exactly the nine permutations
[tp4-CUTLASS],[tp4-TRTLLM],[tp4-TRITON],[ep4-CUTLASS], …,[dp4-TRITON]referenced in the YAML. No changes required.
|
PR_Github #14563 [ run ] completed with state |
|
/bot run --disable-fail-fast --add-multi-gpu-test |
|
PR_Github #14675 [ run ] triggered by Bot |
There was a problem hiding this comment.
Actionable comments posted: 1
🔭 Outside diff range comments (1)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (1)
2556-2563: Bug: LLM kwarg should be moe_config, not moe_backendLLM does not accept a moe_backend kwarg elsewhere in this file; all other usages pass moe_config=MoeConfig(...). This will likely raise a TypeError.
- llm = LLM(self.MODEL_PATH, + llm = LLM(self.MODEL_PATH, tensor_parallel_size=tp_size, pipeline_parallel_size=pp_size, moe_expert_parallel_size=ep_size, kv_cache_config=self.kv_cache_config, **pytorch_config, enable_attention_dp=attention_dp, - moe_backend="TRITON") + moe_config=MoeConfig(backend="TRITON"))
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (3)
tensorrt_llm/evaluate/lm_eval.py(7 hunks)tests/integration/defs/accuracy/test_llm_api_pytorch.py(7 hunks)tests/integration/test_lists/test-db/l0_dgx_h100.yml(1 hunks)
🚧 Files skipped from review as they are similar to previous changes (2)
- tensorrt_llm/evaluate/lm_eval.py
- tests/integration/test_lists/test-db/l0_dgx_h100.yml
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: Python code should conform to Python 3.8+.
Indent Python code with 4 spaces. Do not use tabs.
Always maintain the namespace when importing in Python, even if only one class or function from a module is used.
Python filenames should use snake_case (e.g., some_file.py).
Python classes should use PascalCase (e.g., class SomeClass).
Python functions and methods should use snake_case (e.g., def my_awesome_function():).
Python local variables should use snake_case. Prefix k for variable names that start with a number (e.g., k_99th_percentile).
Python global variables should use upper snake_case and prefix G (e.g., G_MY_GLOBAL).
Python constants should use upper snake_case (e.g., MY_CONSTANT).
Avoid shadowing variables declared in an outer scope in Python.
Initialize all externally visible members of a Python class in the constructor.
For interfaces that may be used outside a Python file, prefer docstrings over comments.
Comments in Python should be reserved for code within a function, or interfaces that are local to a file.
Use Google style docstrings for Python classes and functions, which can be parsed by Sphinx.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the class docstring.
Avoid using reflection in Python when functionality can be easily achieved without it.
When using try-except blocks in Python, limit the except to the smallest set of errors possible.
When using try-except blocks to handle multiple possible variable types in Python, keep the body of the try as small as possible, using the else block to implement the logic.
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
**/*.{cpp,h,hpp,cc,cxx,cu,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the current year. This includes .cpp, .h, .cu, .py, and any other source files which are compiled or interpreted.
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
🧠 Learnings (2)
📚 Learning: 2025-07-28T17:06:08.621Z
Learnt from: moraxu
PR: NVIDIA/TensorRT-LLM#6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
Applied to files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
📚 Learning: 2025-08-06T13:58:07.506Z
Learnt from: galagam
PR: NVIDIA/TensorRT-LLM#6487
File: tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py:1-12
Timestamp: 2025-08-06T13:58:07.506Z
Learning: In TensorRT-LLM, test files (files under tests/ directories) do not require NVIDIA copyright headers, unlike production source code files. Test files typically start directly with imports, docstrings, or code.
Applied to files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Pre-commit Check
🔇 Additional comments (2)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (2)
2488-2494: LLM init path refactor looks goodSwitching to the class-level MODEL_PATH is clean and consistent with other tests.
2525-2533: 4-GPU LLM init matches surrounding patternsConsistent with other tests and uses MoeConfig correctly.
|
PR_Github #14675 [ run ] completed with state |
Signed-off-by: Tracin <10434017+Tracin@users.noreply.github.com>
Signed-off-by: Tracin <10434017+Tracin@users.noreply.github.com>
|
/bot reuse-pipeline |
|
PR_Github #14726 [ reuse-pipeline ] triggered by Bot |
|
PR_Github #14726 [ reuse-pipeline ] completed with state |
3 participants
Summary by CodeRabbit
New Features
Bug Fixes
Tests
Chores
Description
Add GSM8K test case for GPT-OSS.
GSM8K support --fewshot_as_multiturn and return flexible extract scores.
Test Coverage
GitHub Bot Help
/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...Provide a user friendly way for developers to interact with a Jenkins server.
Run
/bot [-h|--help]to print this help message.See details below for each supported subcommand.
run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]Launch build/test pipelines. All previously running jobs will be killed.
--reuse-test (optional)pipeline-id(OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.--disable-reuse-test(OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.--disable-fail-fast(OPTIONAL) : Disable fail fast on build/tests/infra failures.--skip-test(OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.--stage-list "A10-PyTorch-1, xxx"(OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.--gpu-type "A30, H100_PCIe"(OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.--test-backend "pytorch, cpp"(OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.--only-multi-gpu-test(OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.--disable-multi-gpu-test(OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.--add-multi-gpu-test(OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.--post-merge(OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx"(OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".--detailed-log(OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.--debug(OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in thestage-listparameter to access the appropriate container environment. Note: Does NOT update GitHub check status.For guidance on mapping tests to stage names, see
docs/source/reference/ci-overview.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip testing for latest commit on pull request.
--comment "Reason for skipping build/test"is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.reuse-pipeline
reuse-pipelineReuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.