requirements.txt
This is the requirements.txt file for Tensor-RT LLM
Python Library | Explanation |
---|---|
accelerate | A library that simplifies distributed training and mixed precision training for PyTorch models. |
build | A Python library for generating and managing build systems. |
colored | A library for adding colours to terminal output. |
cuda-python | A Python package that provides access to the CUDA API for NVIDIA GPUs. |
diffusers | A library for training and using diffusion models in PyTorch. |
lark | A parsing library for Python that supports context-free grammars. |
mpi4py | A Python wrapper for the Message Passing Interface (MPI) standard, used for parallel computing. |
numpy | A fundamental package for scientific computing in Python, providing support for large, multi-dimensional arrays and matrices. |
onnx | An open format for representing machine learning models, allowing interoperability between different frameworks. |
polygraphy | A library and set of tools for debugging and profiling deep learning models. |
psutil | A cross-platform library for retrieving information on running processes and system utilization. |
pynvml | A Python wrapper for the NVIDIA Management Library (NVML), used for monitoring NVIDIA GPUs. |
pulp | A linear programming modeler written in Python. |
pandas | A data manipulation and analysis library for Python, providing data structures like DataFrames and Series. |
h5py | A Python interface for the HDF5 file format, used for storing large amounts of numerical data. |
StrEnum | A Python enumeration library that allows string values as members. |
sentencepiece | A library for unsupervised text tokenization, used in natural language processing tasks. |
nvidia-cudnn-cu12 | The NVIDIA CUDA Deep Neural Network library (cuDNN) for CUDA 12. |
tensorrt | A high-performance deep learning inference optimiser and runtime engine for production deployment. |
torch | The PyTorch deep learning framework, used for building and training neural networks. |
nvidia-ammo | The NVIDIA PyTorch Acceleration Library (AMMO), a collection of tools for optimising PyTorch models. |
transformers | A library for natural language processing, providing pre-trained models and tools for tasks like text classification and generation. |
wheel | A built-package format for Python, used for distributing and installing Python packages. |
optimum | A library for optimizing and fine-tuning pre-trained models in PyTorch. |
evaluate | A library for evaluating machine learning models, providing metrics and utility functions. |
janus | A Python library for asynchronous programming, providing a thread-safe queue implementation. |
setuptools | A library for packaging and distributing Python projects, used for building and installing packages. |
mpmath | A library for arbitrary-precision floating-point arithmetic in Python. |
Last updated