fix CI warnings when PR

Signed-off-by: Jerry Guan <jerryguan777@gmail.com>

Author: Jerryguan777

examples/evaluation_and_profiling/swe_bench/src/nat_swe_bench/config.py

@@ -21,23 +21,40 @@
 
 from nat.data_models.common import BaseModelRegistryTag
 from nat.data_models.common import TypedBaseModel
-from nat.data_models.component_ref import FunctionRef
 from nat.data_models.component_ref import LLMRef
 from nat.data_models.function import FunctionBaseConfig
 
 
 class SweBenchPredictorBaseConfig(TypedBaseModel, BaseModelRegistryTag):
+    """Base configuration class for SWE-bench predictors."""    
     description: str = "Swe Bench Problem Solver"
 
 
 class SweBenchPredictorGoldConfig(SweBenchPredictorBaseConfig, name="gold"):
+    """Configuration for the gold predictor that uses the provided patch directly.
+    
+    Attributes:
+        verbose: Whether to enable verbose output for debugging.
+    """
     verbose: bool = True
 
 
 class SweBenchPredictorSkeletonConfig(SweBenchPredictorBaseConfig, name="skeleton"):
+    """Configuration for the skeleton predictor template.
+    
+    Attributes:
+        verbose: Whether to enable verbose output for debugging.
+    """
     verbose: bool = False
 
 class SweBenchPredictorIterativeConfig(SweBenchPredictorBaseConfig, name="iterative"):
+    """Configuration for the iterative predictor that solves problems step-by-step.
+    
+    Attributes:
+        llm_name: Reference to the LLM to use for iterative problem solving.
+        step_limit: Maximum number of agent steps before termination.
+        timeout: Command execution timeout in seconds.
+    """    
     llm_name: LLMRef = Field(description="LLM to use for iterative agent")
     step_limit: int = Field(default=250, description="Maximum number of agent steps")
     timeout: int = Field(default=60, description="Command execution timeout in seconds")
@@ -49,4 +66,9 @@ class SweBenchPredictorIterativeConfig(SweBenchPredictorBaseConfig, name="iterat
     Discriminator(TypedBaseModel.discriminator)]
 
 class SweBenchWorkflowConfig(FunctionBaseConfig, name="swe_bench"):
+    """Configuration for the SWE-bench workflow.
+    
+    Attributes:
+        predictor: The predictor configuration (gold, skeleton, or iterative).
+    """
     predictor: SweBenchPredictorConfig

examples/evaluation_and_profiling/swe_bench/src/nat_swe_bench/configs/config_iterative.yml

@@ -1,3 +1,25 @@
+# SPDX-FileCopyrightText: Copyright (c) 2025-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+llms:
+ nim_llm:
+   _type: nim
+   model_name: mistralai/mistral-nemotron
+   temperature: 0.6
+   max_tokens: 4096    
+
 llms:
   claude_sonnet_llm:
     _type: litellm
@@ -12,14 +34,6 @@ llms:
 #     temperature: 0.0
 #     api_key: "${OPENAI_API_KEY}"  # Set this environment variable before running
 
-
-# llms:
-#  nim_llm:
-#    _type: nim
-#    model_name: meta/llama-3.3-70b-instruct
-#    temperature: 0.6
-#    max_tokens: 4096    
-
 workflow:
   _type: swe_bench
   predictor:
@@ -49,18 +63,15 @@ eval:
           field:
             instance_id:
               - sympy__sympy-20590
-              #- sympy__sympy-21055
-              #- sympy__sympy-11400
-              #- sympy__sympy-11870
-              #- astropy__astropy-12907
-              #- astropy__astropy-6938
-              #- django__django-15781
-              #- django__django-11001
-              #- matplotlib__matplotlib-25332
-              #- mwaskom__seaborn-3010
-              #- pallets__flask-4045
-              #- psf__requests-1963
-              #- pydata__xarray-3364
+              # - sympy__sympy-21055
+              # - sympy__sympy-11400
+              # - astropy__astropy-12907
+              # - astropy__astropy-6938
+              # - django__django-15781
+              # - django__django-11001
+              # - mwaskom__seaborn-3010
+              # - pallets__flask-4045
+              # - psf__requests-1963
 
   evaluators:
     swe_bench:

examples/evaluation_and_profiling/swe_bench/src/nat_swe_bench/predictors/predict_iterative/tools/git_tool.py

@@ -1,4 +1,4 @@
-# SPDX-FileCopyrightText: Copyright (c) 2025-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-FileCopyrightText: Copyright (c) 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -13,6 +13,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import asyncio
 import logging
 from dataclasses import dataclass
 from pathlib import Path
@@ -69,17 +70,23 @@ async def cleanup(self):
 
 def get_repo_path(workspace_dir: str, repo_url: str) -> Path:
     """Generate a unique path for the repository."""
-    repo_name = repo_url.split('/')[-1].replace('.git', '')
-    return Path(workspace_dir) / repo_name
+    parts = repo_url.rstrip('/').split('/')
+    repo_name = parts[-1].replace('.git', '')
+    org_name = parts[-2]  # Organization name
+    
+    # Return: workspace_dir/org/repo
+    return Path(workspace_dir) / org_name / repo_name
 
 
 async def clone_repository(repo_url: str, target_path: Path) -> Repo:
     """Clone a repository to the specified path."""
     logger.info("Cloning repository %s to %s", repo_url, target_path)
-    return Repo.clone_from(repo_url, target_path)
+    # Use asyncio.to_thread to avoid blocking the event loop during clone operation
+    return await asyncio.to_thread(Repo.clone_from, repo_url, target_path)
 
 
 async def checkout_commit(repo: Repo, commit_hash: str):
     """Checkout a specific commit in the repository."""
     logger.info("Checking out commit %s", commit_hash)
-    repo.git.checkout(commit_hash)
+    # Use asyncio.to_thread to avoid blocking the event loop during checkout
+    await asyncio.to_thread(repo.git.checkout, commit_hash)

examples/evaluation_and_profiling/swe_bench/src/nat_swe_bench/predictors/predict_iterative/tools/register.py

@@ -1,4 +1,4 @@
-# SPDX-FileCopyrightText: Copyright (c) 2025-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-FileCopyrightText: Copyright (c) 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -39,18 +39,24 @@ async def git_repo_tool(tool_config: GitRepoToolConfig, builder: Builder):
 
     # Simple async function that accepts a JSON string
     async def git_operations(args_str: str) -> str:
-        args = json.loads(args_str)
+        try:
+            args = json.loads(args_str)
+        except json.JSONDecodeError as e:
+            raise ValueError(f"Invalid JSON input: {e}") from e
+
         operation = args.get('operation')
 
         if operation == "setup":
+            if 'repo_url' not in args or 'base_commit' not in args:
+                raise ValueError("setup operation requires 'repo_url' and 'base_commit'")
             context = await repo_manager.setup_repository(args['repo_url'], args['base_commit'])
             return str(context.repo_path)
 
         if operation == "cleanup":
             await repo_manager.cleanup()
             return "Cleanup complete"
 
-        raise ValueError(f"Unknown operation: {operation}")
+        raise ValueError(f"Unknown operation: {operation}. Supported: 'setup', 'cleanup'")
 
     try:
         yield FunctionInfo.from_fn(git_operations,