[BUG] Stops after Neither CUDA nor MPS are available - defaulting to CPU. Note: This module is much faster with a GPU.

[caarson]

Found a bug? Please fill out the sections below. 👍

Describe the bug

After creating an isolated environment in Anaconda using Python 3.11 (I think, whatever worked I used) and typing operate, I would have it operate for a few minutes, however it stops when it gets to Neither CUDA nor MPS are available - defaulting to CPU. Note: This module is much faster with a GPU.

Steps to Reproduce

1. activated the conda environment
2. navigated to the project directory
3. ran operate

Expected Behavior

Would carry on until it confirmed it has completed the task as asked.

Actual Behavior:

`[Self-Operating Computer | gpt-4-with-ocr]
Hello, I can help you with anything. What would you like done?
[User]
Find the location of C:\ProgramData\Docker\ on this computer. I don't believe it is in that particular directory so if
it's not, find out where it is.
[Self-Operating Computer | gpt-4-with-ocr]
I need to search for File Explorer to navigate to the specified directory.
Action: press ['win']

[Self-Operating Computer | gpt-4-with-ocr]
Now I will type 'File Explorer' to search for it.
Action: write File Explorer

[Self-Operating Computer | gpt-4-with-ocr]
I will press enter to open File Explorer.
Action: press ['enter']

[Self-Operating Computer | gpt-4-with-ocr]
Now that File Explorer is open, I need to navigate to the C:\ProgramData\Docker directory.
Action: write C:\ProgramData\Docker

[Self-Operating Computer | gpt-4-with-ocr]
I will press enter to navigate to the directory.
Action: press ['enter']

(CPU JUMPS TO 100%)

Neither CUDA nor MPS are available - defaulting to CPU. Note: This module is much faster with a GPU.
D:\CondaEnvs\soc\Lib\site-packages\easyocr\detection.py:78: FutureWarning: You are using torch.load with weights_only=False (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for weights_only will be flipped to True. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via torch.serialization.add_safe_globals. We recommend you start setting weights_only=True for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
net.load_state_dict(copyStateDict(torch.load(trained_model, map_location=device)))
D:\CondaEnvs\soc\Lib\site-packages\easyocr\recognition.py:169: FutureWarning: You are using torch.load with weights_only=False (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for weights_only will be flipped to True. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via torch.serialization.add_safe_globals. We recommend you start setting weights_only=True for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
state_dict = torch.load(model_path, map_location=device)

(D:\CondaEnvs\soc) D:\soc\Project\self-operating-computer>`

Environment

• OS: Windows 11, isolated Anaconda environment
• Model Used GPT 4 with OCR
• Framework Version (optional):

Screenshots

Additional context

Using a version of Torch for CPU only. Not sure if there are other modules requiring a dedicated GPU. Using a 3rd gen Intel Core i5 with no dedicated graphics card, just integrated iGPU.

[tomahawk000]

Ditto. And I have a RTX A4500