PyTorch - TONYLABS TECH CO., LTD.

What is PyTorch?

PyTorch is an open-source machine learning library developed by Facebook's AI Research lab. It is based on the Torch library and is used for applications such as natural language processing. PyTorch provides two high-level features:

Tensor computation (like NumPy) with strong GPU acceleration
Deep neural networks built on a tape-based autograd system

PyTorch is widely used in the research community and is known for its flexibility and ease of use.

Installation

Check out the latest version of PyTorch installation instructions on the official website.

It's recommended to create a virtual environment before installing PyTorch to avoid conflicts with other Python packages.

Linux

Platform	CUDA Version	Command
Linux	None	`pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu`
Linux	CUDA 13.0	`pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu130`
Linux	CUDA 12.8	`pip3 install torch torchvision torchaudio`
Linux	CUDA 12.6	`pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126`
MacOS	None	`pip3 install torch torchvision torchaudio`
Windows	None	`pip3 install torch torchvision torchaudio`
Windows	CUDA 12.6	`pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126`

Verify Installation

To make sure PyTorch is installed correctly, you can run the following Python code:

import torch
x = torch.rand(5, 3)
print(x)

Output:

tensor([[0.3380, 0.3845, 0.3217],
       [0.8337, 0.9050, 0.2650],
       [0.2979, 0.7141, 0.9069],
       [0.1449, 0.1132, 0.1375],
       [0.4675, 0.3947, 0.1426]])

NVIDIA CUDA Toolkit

Learn more about CUDA Toolkit.

Checks in PyTorch to validate whether a GPU (CUDA-capable NVIDIA device) is available and usable.

import torch

gpu_count = torch.cuda.device_count()
print(f"Number of GPUs: {gpu_count}")

for i in range(gpu_count):
    print(f"GPU {i}: {torch.cuda.get_device_name(i)}")