lzy/sci-gui-agent-benchmark
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Feat/monitor (#254)

Browse Source 
* feat: add claude support

* feat: add script for end-to-end evaluation with logging and task distribution

* feat&fix: add tool result handling and update model default in evaluation script

* chore: remove run_test_env.py script

* feat&fix: implement action parsing for tool calls and update default action space

* fix: update text formatting in action parsing and replace logger import

* feat&fix: implement action parsing for tool calls and add screen size handling

* feat: add setup instructions for Anthropic API integration

* feat: add notice about image size limitations for Anthropic API

* Delete test_env/logger.py

* Delete test_env/utils.py

* fix: update logger usage to use global logger and improve error handling

* feat&fix: add configuration management API endpoints and update UI for configuration selection

* feat&fix: update environment configuration, enhance task statistics, and improve UI responsiveness

* feat&fix: add configuration toggle button in UI and improve task loading performance

* feat&fix: add accuracy percentage display to score and style updates for UI
This commit is contained in:
Zilong Zhou
2025-07-14 13:43:41 +08:00
committed by GitHub
parent 0651495d88
commit 74b7c189af
6 changed files with 662 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
mm_agents/anthropic/main.py
View File

@@ -369,9 +369,10 @@ class AnthropicAgent:
)
except (APIError, APIStatusError, APIResponseValidationError) as e:
self.logger.exception(f"Anthropic API error: {str(e)}")
logger.exception(f"Anthropic API error: {str(e)}")
try:
self.logger.warning("Retrying with backup API key...")
logger.warning("Retrying with backup API key...")
backup_client = Anthropic(api_key=os.environ.get("ANTHROPIC_API_KEY_BACKUP"), max_retries=4)
if self.model_name == "claude-3-7-sonnet-20250219" or self.model_name == "claude-4-opus-20250514" or self.model_name == "claude-4-sonnet-20250514":
@@ -393,13 +394,13 @@ class AnthropicAgent:
tools=tools,
betas=betas,
)
self.logger.info("Successfully used backup API key")
logger.info("Successfully used backup API key")
except Exception as backup_e:
self.logger.exception(f"Backup API call also failed: {str(backup_e)}")
logger.exception(f"Backup API call also failed: {str(backup_e)}")
return None, None
except Exception as e:
self.logger.exception(f"Error in Anthropic API: {str(e)}")
logger.exception(f"Error in Anthropic API: {str(e)}")
return None, None
response_params = _response_to_params(response)
@@ -434,9 +435,15 @@ class AnthropicAgent:
actions = ["DONE"]
return reasonings, actions
def reset(self, *args, **kwargs):
def reset(self, _logger = None, *args, **kwargs):
"""
Reset the agent's state.
"""
global logger
if _logger:
logger = _logger
else:
logger = logging.getLogger("desktopenv.agent")
self.messages = []
self.logger.info(f"{self.class_name} reset.")
logger.info(f"{self.class_name} reset.")