[TRTLLM-6823][doc] Add checkpoint refactor docs #6592
Conversation
Signed-off-by: Shahar Mor <17088876+shaharmor98@users.noreply.github.com>
📝 WalkthroughWalkthroughA new documentation file was added describing the modular checkpoint loading system for PyTorch in TRTLLM, including usage and customization. Additionally, a typo in the method name Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant LLM
participant CheckpointLoader
participant ConfigLoader
participant WeightLoader
participant WeightMapper
User->>LLM: load_model(checkpoint_dir)
LLM->>CheckpointLoader: load(checkpoint_dir)
CheckpointLoader->>ConfigLoader: load(checkpoint_dir)
ConfigLoader-->>CheckpointLoader: ModelConfig
CheckpointLoader->>WeightLoader: load_weights(checkpoint_dir)
WeightLoader-->>CheckpointLoader: weights_dict
CheckpointLoader->>WeightMapper: map_weights(model, weights_dict)
WeightMapper-->>CheckpointLoader: mapped_weights
CheckpointLoader-->>LLM: loaded_model
Estimated code review effort🎯 2 (Simple) | ⏱️ ~7 minutes 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. ✨ 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
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (2)
docs/source/torch/features/checkpoint_loading.md (2)
116-116: Minor wording tweak for clarity“…provide any specific subcomponent upon need” → “…provide any specific subcomponent when needed”.
- checkpoint_loader=HfCheckpointLoader(weight_loader=MyCustomWeightLoader())) + checkpoint_loader=HfCheckpointLoader(weight_loader=MyCustomWeightLoader())) -... they can provide any specific subcomponent upon need +... they can provide any specific subcomponent when needed
332-332: Spelling typo: “asscoiated” → “associated”-By setting the model name, the registered mapper will be asscoiated with the specific model. +By setting the model name, the registered mapper will be associated with the specific model.
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (3)
docs/source/torch/features/checkpoint_loading.md(1 hunks)tensorrt_llm/_torch/models/checkpoints/base_checkpoint_loader.py(1 hunks)tensorrt_llm/_torch/pyexecutor/model_engine.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: The code developed for TensorRT-LLM 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 file, prefer docstrings over comments in Python.
Comments in Python should be reserved for code within a function, or interfaces that are local to a file.
Use Google style docstrings for classes and functions in Python, which can be parsed by Sphinx.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the docstring for the class.
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/_torch/models/checkpoints/base_checkpoint_loader.pytensorrt_llm/_torch/pyexecutor/model_engine.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/_torch/models/checkpoints/base_checkpoint_loader.pytensorrt_llm/_torch/pyexecutor/model_engine.py
🧠 Learnings (2)
📓 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: in tensorrt-llm testing, it's common to have both cli flow tests (test_cli_flow.py) and pytorch api ...
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:
docs/source/torch/features/checkpoint_loading.md
🪛 LanguageTool
docs/source/torch/features/checkpoint_loading.md
[style] ~146-~146: Consider using a different adverb to strengthen your wording.
Context: ...ent all four components when supporting a completely new checkpoint format - **Custom Weight...
(COMPLETELY_ENTIRELY)
[grammar] ~332-~332: Ensure spelling is correct
Context: ...del name, the registered mapper will be asscoiated with the specific model.
(QB_NEW_EN_ORTHOGRAPHY_ERROR_IDS_1)
🔇 Additional comments (3)
tensorrt_llm/_torch/models/checkpoints/base_checkpoint_loader.py (1)
70-71: LGTM! Typo correction improves code quality.The method name correction from
get_initilized_weight_mappertoget_initialized_weight_mapperfixes the misspelling and improves code readability. The method signature and functionality remain unchanged.tensorrt_llm/_torch/pyexecutor/model_engine.py (1)
1061-1061: LGTM! Method call updated to match corrected method name.The method call correction from
get_initilized_weight_mappertoget_initialized_weight_mappermaintains consistency with the updated method name in BaseCheckpointLoader. This ensures the code continues to function correctly after the typo fix.docs/source/torch/features/checkpoint_loading.md (1)
35-36: All references to the old misspelling have been removedI ran
rg -n "get_initilized_weight_mapper"across the repository and found no matches—there are no remaining stale references.
nv-guomingz
force-pushed
the
docs/add-checkpoint-docs
branch
from
5a63684 to
c0ca8f0
Compare
|
/bot run |
|
PR_Github #14444 [ run ] triggered by Bot |
|
PR_Github #14444 [ run ] completed with state |
chenopis
left a comment
chenopis
left a comment
There was a problem hiding this comment.
Some grammar and spelling nits. LGTM overall.
|
/bot run |
Signed-off-by: Shahar Mor <17088876+shaharmor98@users.noreply.github.com>
shaharmor98
force-pushed
the
docs/add-checkpoint-docs
branch
from
c0ca8f0 to
d3aa1f9
Compare
There was a problem hiding this comment.
Actionable comments posted: 2
♻️ Duplicate comments (2)
docs/source/torch/features/checkpoint_loading.md (2)
321-323: Missing colon makes the example invalid PythonAdd the trailing colon.
@register_mapper("CUSTOM_FORMAT") -class CustomWeightMapper(BaseWeightMapper) +class CustomWeightMapper(BaseWeightMapper):
328-330: Missing colon in model-specific mapper exampleAdd the trailing colon.
@register_mapper("CUSTOM_FORMAT", "Gemma3ForCausalLM") -class CustomWeightMapper(BaseWeightMapper) +class CustomWeightMapper(BaseWeightMapper):
🧹 Nitpick comments (4)
docs/source/torch/features/checkpoint_loading.md (4)
82-85: Normalize phrasing and casing in the HF features bulletsMake the three bullets parallel and consistently phrased.
- - **Weights loading** (`.safetensors, .bin, .pth`): Load HF-compatible weights from disk - - **Configuration parser** - Parse configuration information stored by HF into a TRTLLM `ModelConfig` object - - **Weights Mapping** - Convert HF weights into a TRTLLM-compatible representation + - **Weight loading** (`.safetensors`, `.bin`, `.pth`): Load HF-compatible weights from disk. + - **Configuration parsing**: Parse configuration information stored by HF into a TRTLLM `ModelConfig` object. + - **Weight mapping**: Convert HF weights into a TRTLLM-compatible representation.
90-90: Add missing periodEnd the sentence with a period.
-There are two main approaches for using checkpoint loading objects +There are two main approaches for using checkpoint loading objects.
63-65: Missing typing import forAnyThe signature uses
dict[str, Any]butAnyisn’t imported in this snippet.+from typing import Any from tensorrt_llm._torch.models.checkpoints.base_weight_loader import BaseWeightLoader
146-146: Grammar: “Completely New Format”Use the adverb form.
-- **Complete New Format**: Implement all four components to support a new checkpoint format +- **Completely New Format**: Implement all four components to support a new checkpoint format
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
docs/source/torch/features/checkpoint_loading.md(1 hunks)
⏰ 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
|
PR_Github #14705 [ run ] triggered by Bot |
|
/bot run --disable-fail-fast |
|
PR_Github #14706 [ run ] triggered by Bot |
|
PR_Github #14705 [ run ] completed with state |
|
PR_Github #14706 [ run ] completed with state |
|
/bot run --disable-fail-fast |
|
PR_Github #14715 [ run ] triggered by Bot |
|
PR_Github #14715 [ run ] completed with state |
8 participants
Summary by CodeRabbit
Documentation
Bug Fixes
get_initilized_weight_mappertoget_initialized_weight_mapperto ensure consistency and prevent potential errors.Description
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.
Details
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.