JSONDecodeError during Code Execution (before executing agent-written code)

[tezansahu]

🐛 Bug Description

Facing the following error for every code execution stage when the general_model is used to implement a paper:

--------------Execution feedback:---------------
Execution error: Extra data: line 2 column 1 (char 77)
Traceback: Traceback (most recent call last):
  File "/mnt/d/TechStuff/RD-Agent/rdagent/components/coder/model_coder/model.py", line 97, in execute
    qtde.prepare()
  File "/mnt/d/TechStuff/RD-Agent/rdagent/utils/env.py", line 374, in prepare
    super().prepare()
  File "/mnt/d/TechStuff/RD-Agent/rdagent/utils/env.py", line 211, in prepare
    status_dict = json.loads(part)
  File "/home/tezansahu/anaconda3/envs/rdagent/lib/python3.10/json/__init__.py", line 346, in loads
    return _default_decoder.decode(s)
  File "/home/tezansahu/anaconda3/envs/rdagent/lib/python3.10/json/decoder.py", line 340, in decode
    raise JSONDecodeError("Extra data", s, end)
json.decoder.JSONDecodeError: Extra data: line 2 column 1 (char 77)

All other model responses like writing the description/formulation, as well as the response of critics on the code are returned correctly.

Command used to run: rdagent general_model --report_file_path=rdagent/scenarios/general_model/1412.6550v4_FitNet.pdf

To Reproduce

Steps to reproduce the behavior:

1. Installed rdagent using instructions for devs: https://rdagent.readthedocs.io/en/stable/development.html
2. Created .env file with the following settings:

   USE_AZURE=True
   USE_AZURE_TOKEN_PROVIDER=False
   MAX_RETRY=10
   RETRY_WAIT_SECONDS=20
   
   OPENAI_API_KEY=<removed>
   CHAT_MODEL=gpt-4o
   CHAT_MAX_TOKENS=3000
   CHAT_TEMPERATURE=0.7
   CHAT_AZURE_API_BASE=<removed>
   CHAT_AZURE_API_VERSION=2024-06-01
   
   EMBEDDING_MODEL=text-embedding-ada-002
   EMBEDDING_AZURE_API_BASE=<removed>
   EMBEDDING_AZURE_API_VERSION=2024-06-01
   

3. Downloaded the FitNet paper & placed it in rdagent/scenarios/general_model/ folder
4. Ran the command rdagent general_model --report_file_path=rdagent/scenarios/general_model/1412.6550v4_FitNet.pdf

Expected Behavior

Code execution to happen successfully & actual code errors to be found (not setup errors)

Screenshot

Environment

Note: Users can run rdagent collect_info to get system information and paste it directly here.

• Name of current operating system: Windows (using WSL 2 - Ubuntu 22.04LTS)
• Processor architecture: AMD64
• System, version, and hardware information: Microsoft Windows 11 Home Single Language (OS Version: 10.0.22631 N/A Build 22631) | Intel64 Family 6 Model 154 Stepping 3 GenuineIntel ~2700 Mhz processor
• Version number of the system: 10.0.22631 N/A Build 22631
• Python version: 3.10.15
• Container ID:
• Container Name:
• Container Status:
• Image ID used by the container:
• Image tag used by the container:
• Container port mapping:
• Container Label:
• Startup Commands: rdagent general_model --report_file_path=rdagent/scenarios/general_model/1412.6550v4_FitNet.pdf
• RD-Agent version: 0.2.1
• Package version:

[SunsetWolf]

The issue you mentioned has been fixed in PR 386, please pull the latest code and run it and the issue will be resolved.