Please praprae this anaconda, CUDA, CuDNN, and HuggingFace AccessToken for practice.
Follow this step!!
- A.1. Install Anacdona for practice
- A.2. Check Graphic Card Name
- A.3. Install NVIDIA DRIVER
- A.4. Check Compatibility
- A.5. Install Python@3.8.16 Depdencies
- A.6. Some modules
When you're trying computer vision, you need to install some dependency, not supported windows.
And then, you can use anaconda.
I used conda 23.1.0.
- Install environment
conda env create -f requirements-*.yml- Export environment
conda env export > requirements-*.yml- Check [Windwos] - [Device Administrators] - [Display Adaptor]
- Disable Basic GPU Instance supplied by CPU.
- Notes
NVIDIA GeForce RTX 3080 Laptop GPUto install NVIDIA Driver
- Install NVIDIDA DRIVER for
NVIDIA GeForce RTX 3080 Laptop GPUor your instances.
-
Check Graphics Card Compatibility among
NVIDIA GeForce RTX 3080 Laptop GPUand CUDA Computing Power
->NVIDIA GeForce RTX 3080supports CUDA Computing Function 8.6 -
Check Python 3.8.16 Compatibility among CUDA and CUDDN
-> Deprecation of CUDA 11.6 and Python 3.7 Support
| PyTorch Version | Python | Stable CUDA | Experimental CUDA |
|---|---|---|---|
| 2.0 | >=3.8, <=3.11 | CUDA 11.7, CUDNN 8.5.0.96 | CUDA 11.8, CUDNN 8.7.0.84 |
| 1.13 | >=3.7, <=3.10 | CUDA 11.6, CUDNN 8.3.2.44 | CUDA 11.7, CUDNN 8.5.0.96 |
| 1.12 | >=3.7, <=3.10 | CUDA 11.3, CUDNN 8.3.2.44 | CUDA 11.6, CUDNN 8.3.2.44 |
- Install environment
pip install -r requirements-pip.txt- Export environment
pip freeze >> requirements-pip.txtSome Module can't install without HuggingFace AccessToken. Please signup and publish AccessToken to Downlaod these files....