multiple calls to text-generation-webui

[mongolu]

Description

When starting webui+backend+frontend (in this order) and in web interface "TASK AGENT" sent text "compose a joke about pirates", it's tarting to send to webui many calls. I don't know if it's webui problem, my error, or Agent-LLM problem, but the output in webui is: "

Traceback (most recent call last): File "/home/mihai/text-generation-webui/modules/text_generation.py", line 272, in generate_reply output = shared.model.generate(**generate_params)[0] File "/home/mihai/.local/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context return func(*args, **kwargs) File "/home/mihai/.local/lib/python3.10/site-packages/transformers/generation/utils.py", line 1276, in generate and torch.sum(inputs_tensor[:, -1] == generation_config.pad_token_id) > 0 IndexError: index -1 is out of bounds for dimension 1 with size 0 Output generated in 0.01 seconds (0.00 tokens/s, 0 tokens, context 0, seed 140062149) 127.0.0.1 - - [03/May/2023 20:25:11] "POST /api/v1/generate HTTP/1.1" 200 -

---

Traceback (most recent call last): File "/home/mihai/text-generation-webui/modules/text_generation.py", line 272, in generate_reply output = shared.model.generate(**generate_params)[0] File "/home/mihai/.local/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context return func(*args, **kwargs) File "/home/mihai/.local/lib/python3.10/site-packages/transformers/generation/utils.py", line 1276, in generate and torch.sum(inputs_tensor[:, -1] == generation_config.pad_token_id) > 0 IndexError: index -1 is out of bounds for dimension 1 with size 0 Output generated in 0.05 seconds (0.00 tokens/s, 0 tokens, context 0, seed 793370993) 127.0.0.1 - - [03/May/2023 20:25:11] "POST /api/v1/generate HTTP/1.1" 200 - ╭─────────────────────────────── Traceback (most recent call last) ────────────────────────────────╮ │ /home/mihai/text-generation-webui/server.py:929 in │ │ │ │ 926 │ # Launch the web UI │ │ 927 │ create_interface() │ │ 928 │ while True: │ │ ❱ 929 │ │ time.sleep(0.5) │ │ 930 │ │ if shared.need_restart: │ │ 931 │ │ │ shared.need_restart = False │ │ 932 │ │ │ shared.gradio['interface'].close() │ ╰──────────────────────────────────────────────────────────────────────────────────────────────────╯ KeyboardInterrupt " It seems to me that no text is sent on the interface API of webui.

Steps to Reproduce the Bug

1. start text-generation-webui
2. start backend
3. start frontend
4. go to webpage http://localhost:3000/agent/XXX to TASK AGENT
5. send text "compose a joke about pirates"

Expected Behavior

to function like in previous versions, i presume like in v1.0.17-alpha

Actual Behavior

It's starting a subsecond comunication with webui, which result is always like this: "Response:

New Tasks:

[{'task_name': ''}]

Response:

Executing task 1: Develop a task list.

Response:

Task Result:

Response:

New Tasks:

[{'task_name': ''}]

Response:

Executing task 1: Develop a task list.

Response:

Task Result: "

Additional Context / Screenshots

No response

Operating System

• [ ] Microsoft Windows
• [ ] Apple MacOS
• [ ] Linux
• [ ] Android
• [ ] iOS
• [X] Other

Python Version

• [ ] Python <= 3.9
• [X] Python 3.10
• [ ] Python 3.11

Environment Type - Connection

• [X] Local
• [ ] Remote

Environment Type - Container

• [ ] Using Docker
• [X] Not Using Docker

Acknowledgements

• [X] My issue title is concise, descriptive, and in title casing.
• [X] I have searched the existing issues to make sure this bug has not been reported yet.
• [X] I am using the latest version of Agent-LLM.
• [X] I have provided enough information for the maintainers to reproduce and diagnose the issue.