Not able to launch demo app

[Sandrinotje213]

Hello, I wanted to try out AmadeusGPT but I do not have much experience with projects like this. Perhaps someone can help me to launch the app, as I feel I am missing something that is obvious.

I have followed the installation page, using Anaconda Prompt (terminal). I created an environment (although I had to remove the apostrophes in the pip install 'amadeusgpt[streamlit]' command), gitcloned the demo files, and went to the correct directory. No errors appear, until I try to actually launch the app. The terminal does not recognise ' make app' as a command, nor did ' python -m launch_amadeusGPT' work. The Anaconda terminal then returns the following:

---

_Traceback (most recent call last):
File "C:\ProgramData\Anaconda3\lib\runpy.py", line 197, in _run_module_as_main
return run_code(code, main_globals, None,
File "C:\ProgramData\Anaconda3\lib\runpy.py", line 87, in run_code
exec(code, run_globals)
File "C:\Users\DevNeurosci\Documents\AmadeusGPT\launch_amadeusGPT.py", line 3, in
from amadeusgpt import app
File "C:\Users\DevNeurosci\Documents\AmadeusGPT\amadeusgpt_init.py", line 8, in
from amadeusgpt.implementation import AnimalBehaviorAnalysis
File "C:\Users\DevNeurosci\Documents\AmadeusGPT\amadeusgpt\implementation.py", line 8, in
from amadeusgpt import utils
File "C:\Users\DevNeurosci\Documents\AmadeusGPT\amadeusgpt\utils.py", line 12, in
from amadeusgpt.logger import AmadeusLogger
File "C:\Users\DevNeurosci\Documents\AmadeusGPT\amadeusgpt\logger.py", line 1, in
from google.cloud import firestore
ModuleNotFoundError: No module named 'google'

---

I also tried to launch the app in Google Colab, following the 'examples[simple_amadeusGPT_demo.ipynb' file. I gitcloned the Amadeus GitHub files here as well, altered the file locations to fit the Google Colab. Everything works until I want to run "answer = AMADEUS.chat_iteration(query)". Colab then returns the following:

---

_2024-04-19 08:17:12.938
Warning: to view this Streamlit app on a browser, run it with the following
command:

streamlit run /usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py [ARGUMENTS]

---

TypeError Traceback (most recent call last)
in <cell line: 1>()
----> 1 answer = AMADEUS.chat_iteration(query)
2 render(answer)

1 frames
/usr/local/lib/python3.10/dist-packages/amadeusgpt/main.py in load_module_smartly(cls, user_input)
126 def load_module_smartly(cls, user_input):
127 sorted_query_results = match_module(user_input)
--> 128 if len(sorted_query_results) == 0:
129 return None
130 # query result sorted by most relevant module text

TypeError: object of type 'NoneType' has no len()_

---

I then attempted to run !streamlit run /usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py [query] after defining 'query' , but this seems to just create a loop (waited for 5 minutes) after saying the connection was succesful (IPs are masked):

---

_Collecting usage statistics. To deactivate, set browser.gatherUsageStats to false.

You can now view your Streamlit app in your browser.

Network URL: http://[ip1]/
External URL: http://[ip2]/_

---

Any help would be appreciated.

[yeshaokai]

@Sandrinotje213

Thanks for reporting the issue! We are testing a new release that should fix this and it's much better than the previous one. Feel free to test it and give us feedback!

https://github.com/AdaptiveMotorControlLab/AmadeusGPT/tree/release-0.1.0

[laurelrr]

Hello,
I thought I would chime in here but let me know if you prefer me to open an issue. I have just installed amadeusgpt through conda and exported my open ai key as instructed in the conda environment. I am trying to run the jupyter notebook EPM_demo and I get a message not able to import cebra right after I run the first cell. Is this a dependency that I should install separately?

But the main error comes after the "Set the scene number..." cell:

Project created at results. Results will be saved to results
The project will load video files (*.mp4) and optionally keypoint files from ../examples/EPM
A copy of the project config file is saved at results/config.yaml
{'data_info': {'data_folder': '../examples/EPM',
               'result_folder': 'results',
               'video_suffix': '.mp4'},
 'keypoint_info': {'body_orientation_keypoints': {'animal_center': 'neck',
                                                  'neck': 'nose',
                                                  'tail_base': 'tail_base'},
                   'head_orientation_keypoints': {'neck': 'neck',
                                                  'nose': 'nose'},
                   'include_confidence': False,
                   'use_3d': False},
 'llm_info': {'gpt_model': 'gpt-4o',
              'keep_last_n_messages': 2,
              'max_tokens': 4096,
              'temperature': 0.0},
 'object_info': {'load_objects_from_disk': False, 'use_grid_objects': False},
 'video_info': {'scene_frame_number': 400}}
INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 404 Not Found"
INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 404 Not Found"
error Traceback (most recent call last):
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/amadeusgpt/analysis_objects/llm.py", line 101, in connect_gpt_oai_1
    response = client.chat.completions.create(**json_data)
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_utils/_utils.py", line 275, in wrapper
    return func(*args, **kwargs)
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/resources/chat/completions.py", line 829, in create
    return self._post(
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_base_client.py", line 1280, in post
    return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_base_client.py", line 957, in request
    return self._request(
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_base_client.py", line 1061, in _request
    raise self._make_status_error_from_response(err.response) from None
openai.NotFoundError: Error code: 404 - {'error': {'message': 'The model `gpt-4o` does not exist or you do not have access to it.', 'type': 'invalid_request_error', 'param': None, 'code': 'model_not_found'}}

error Traceback (most recent call last):
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/amadeusgpt/analysis_objects/llm.py", line 101, in connect_gpt_oai_1
    response = client.chat.completions.create(**json_data)
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_utils/_utils.py", line 275, in wrapper
    return func(*args, **kwargs)
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/resources/chat/completions.py", line 829, in create
    return self._post(
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_base_client.py", line 1280, in post
    return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_base_client.py", line 957, in request
    return self._request(
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_base_client.py", line 1061, in _request
    raise self._make_status_error_from_response(err.response) from None
openai.NotFoundError: Error code: 404 - {'error': {'message': 'The model `gpt-4o` does not exist or you do not have access to it.', 'type': 'invalid_request_error', 'param': None, 'code': 'model_not_found'}}

INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 404 Not Found"
error Traceback (most recent call last):
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/amadeusgpt/analysis_objects/llm.py", line 101, in connect_gpt_oai_1
    response = client.chat.completions.create(**json_data)
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_utils/_utils.py", line 275, in wrapper
    return func(*args, **kwargs)
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/resources/chat/completions.py", line 829, in create
    return self._post(
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_base_client.py", line 1280, in post
    return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_base_client.py", line 957, in request
    return self._request(
  File "/home/lkeyes/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/openai/_base_client.py", line 1061, in _request
    raise self._make_status_error_from_response(err.response) from None
openai.NotFoundError: Error code: 404 - {'error': {'message': 'The model `gpt-4o` does not exist or you do not have access to it.', 'type': 'invalid_request_error', 'param': None, 'code': 'model_not_found'}}

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Cell In[3], line 17
      4 kwargs = {
      5     "keypoint_info.body_orientation_keypoints.neck" : "nose",
      6     "keypoint_info.body_orientation_keypoints.tail_base" : "tail_base",
   (...)
     10     "video_info.scene_frame_number" : scene_frame_number,
     11 }
     13 config = create_project(data_folder = "../examples/EPM",
     14                         result_folder = "results",
     15                         **kwargs
     16                         )
---> 17 amadeus = AMADEUS(config, use_vlm=True)
     18 video_file_paths = amadeus.get_video_file_paths()
     19 print (video_file_paths)

File ~/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/amadeusgpt/main.py:64, in AMADEUS.__init__(self, config, use_vlm, movie2keypoint_func)
     62 # can only do this after the register process
     63 if use_vlm:
---> 64     self.sandbox.configure_using_vlm()

File ~/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/amadeusgpt/programs/sandbox.py:227, in Sandbox.configure_using_vlm(self)
    224 for identifier, analysis in self.analysis_dict.items():
    226     scene_image = analysis.visual_manager.get_scene_image()
--> 227     json_obj = self.llms["visual_llm"].speak(self, scene_image)
    228     self.meta_info[identifier] = json_obj
    229     # configure meta info on the analysis managers

File ~/anaconda3/envs/amadeusgpt/lib/python3.10/site-packages/amadeusgpt/analysis_objects/llm.py:263, in VisualLLM.speak(self, sandbox, image)
    256 self.update_history(
    257     "user",
    258     "here is the image",
    259     multi_image_content=multi_image_content,
    260     in_place=True,
    261 )
    262 response = self.connect_gpt(self.context_window, max_tokens=2000)
--> 263 text = response.choices[0].message.content.strip()
    265 print("description of the image frame provided")
    266 print(text)

AttributeError: 'NoneType' object has no attribute 'choices'

[laurelrr]

note: version is 0.1.3

[laurelrr]

also, minor typo in your link to the notebooks We provide example notebooks [here](https://github.com/AdaptiveMotorControlLab/AmadeusGPT/blob/main/notebook)!
this seems to have changed