Skip to content

Latest commit

 

History

History
executable file
·
179 lines (123 loc) · 8.66 KB

README.md

File metadata and controls

executable file
·
179 lines (123 loc) · 8.66 KB
Raw

bitsandbytes-rocm

Credits/History:

The library was tested inside the Docker ROCm/PyTorch:latest container image - Ubuntu 20.04/ROCm 5.0.0/Python3.8:

docker run --net=host -it --device=/dev/kfd --device=/dev/dri rocm/pytorch:latest # NOTE: /dev/kfd and /dev/dri must be passed into Docker for ROCm to work

Pre-Requisites

  • An AMD GPU capable of supporting ROCm and an appropriate amdgpu driver

  • Assumes your ROCm tools are installed in /opt/rocm/ and rocminfo is in your path (often found in /opt/rocm/bin/), if not edit the Makefile to match your distro and/or update your PATH before running.

Compiling

# activate your VENV, if using this within a VENV
git clone https://github.com/deftdawg/bitsandbytes-rocm
export CUDA_VERSION=$(rocminfo | grep -oE "gfx.*" | grep -v gfx000 | sort -u -t 'x' -k 2n | tail -1) # Use ROCm tools to auto detect card
make hip
python setup.py install
python3 -m bitsandbytes # to validate it works

bitsandbytes

The bitsandbytes is a lightweight wrapper around CUDA custom functions, in particular 8-bit optimizers, matrix multiplication (LLM.int8()), and quantization functions.

Resources:

TL;DR

Requirements Python >=3.8. Linux distribution (Ubuntu, MacOS, etc.) + CUDA > 10.0. LLM.int8() requires Turing or Ampere GPUs.

Installation: pip install bitsandbytes

Using 8-bit optimizer:

  1. Comment out optimizer: #torch.optim.Adam(....)
  2. Add 8-bit optimizer of your choice bnb.optim.Adam8bit(....) (arguments stay the same)
  3. Replace embedding layer if necessary: torch.nn.Embedding(..) -> bnb.nn.Embedding(..)

Using 8-bit Inference:

  1. Comment out torch.nn.Linear: #linear = torch.nn.Linear(...)
  2. Add bnb 8-bit linear light module: linear = bnb.nn.Linear8bitLt(...) (base arguments stay the same)
  3. There are two modes:
    • Mixed 8-bit training with 16-bit main weights. Pass the argument has_fp16_weights=True (default)
    • Int8 inference. Pass the argument has_fp16_weights=False
  4. To use the full LLM.int8() method, use the threshold=k argument. We recommend k=6.0.
# LLM.int8()
linear = bnb.nn.Linear8bitLt(dim1, dim2, bias=True, has_fp16_weights=False, threshold=6.0)
# inputs need to be fp16
out = linear(x.to(torch.float16))

Features

  • 8-bit Matrix multiplication with mixed precision decomposition
  • LLM.int8() inference
  • 8-bit Optimizers: Adam, AdamW, RMSProp, LARS, LAMB (saves 75% memory)
  • Stable Embedding Layer: Improved stability through better initialization, and normalization
  • 8-bit quantization: Quantile, Linear, and Dynamic quantization
  • Fast quantile estimation: Up to 100x faster than other algorithms

Requirements & Installation

Requirements: anaconda, cudatoolkit, pytorch

Hardware requirements:

  • LLM.int8(): NVIDIA Turing (RTX 20xx; T4) or Ampere GPU (RTX 30xx; A4-A100); (a GPU from 2018 or older).
  • 8-bit optimizers and quantization: NVIDIA Kepler GPU or newer (>=GTX 78X).

Supported CUDA versions: 10.2 - 12.0

The bitsandbytes library is currently only supported on Linux distributions. Windows is not supported at the moment.

The requirements can best be fulfilled by installing pytorch via anaconda. You can install PyTorch by following the "Get Started" instructions on the official website.

To install run:

pip install bitsandbytes

Using bitsandbytes

Using Int8 Matrix Multiplication

For straight Int8 matrix multiplication with mixed precision decomposition you can use bnb.matmul(...). To enable mixed precision decomposition, use the threshold parameter:

bnb.matmul(..., threshold=6.0)

For instructions how to use LLM.int8() inference layers in your own code, see the TL;DR above or for extended instruction see this blog post.

Using the 8-bit Optimizers

With bitsandbytes 8-bit optimizers can be used by changing a single line of code in your codebase. For NLP models we recommend also to use the StableEmbedding layers (see below) which improves results and helps with stable 8-bit optimization. To get started with 8-bit optimizers, it is sufficient to replace your old optimizer with the 8-bit optimizer in the following way:

import bitsandbytes as bnb

# adam = torch.optim.Adam(model.parameters(), lr=0.001, betas=(0.9, 0.995)) # comment out old optimizer
adam = bnb.optim.Adam8bit(model.parameters(), lr=0.001, betas=(0.9, 0.995)) # add bnb optimizer
adam = bnb.optim.Adam(model.parameters(), lr=0.001, betas=(0.9, 0.995), optim_bits=8) # equivalent


torch.nn.Embedding(...) ->  bnb.nn.StableEmbedding(...) # recommended for NLP models

Note that by default all parameter tensors with less than 4096 elements are kept at 32-bit even if you initialize those parameters with 8-bit optimizers. This is done since such small tensors do not save much memory and often contain highly variable parameters (biases) or parameters that require high precision (batch norm, layer norm). You can change this behavior like so:

# parameter tensors with less than 16384 values are optimized in 32-bit
# it is recommended to use multiplies of 4096
adam = bnb.optim.Adam8bit(model.parameters(), min_8bit_size=16384)

Change Bits and other Hyperparameters for Individual Parameters

If you want to optimize some unstable parameters with 32-bit Adam and others with 8-bit Adam, you can use the GlobalOptimManager. With this, we can also configure specific hyperparameters for particular layers, such as embedding layers. To do that, we need two things: (1) register the parameter while they are still on the CPU, (2) override the config with the new desired hyperparameters (anytime, anywhere). See our guide for more details

Fairseq Users

To use the Stable Embedding Layer, override the respective build_embedding(...) function of your model. Make sure to also use the --no-scale-embedding flag to disable scaling of the word embedding layer (nor replaced with layer norm). You can use the optimizers by replacing the optimizer in the respective file (adam.py etc.).

Release and Feature History

For upcoming features and changes and full history see Patch Notes.

Errors

  1. RuntimeError: CUDA error: no kernel image is available for execution on the device. Solution
  2. _fatbinwrap.. Solution

Compile from source

To compile from source, please follow the compile_from_source.md instructions.

License

The majority of bitsandbytes is licensed under MIT, however portions of the project are available under separate license terms: Pytorch is licensed under the BSD license.

We thank Fabio Cannizzo for his work on FastBinarySearch which we use for CPU quantization.

How to cite us

If you found this library and found LLM.int8() useful, please consider citing our work:

@article{dettmers2022llmint8,
  title={LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale},
  author={Dettmers, Tim and Lewis, Mike and Belkada, Younes and Zettlemoyer, Luke},
  journal={arXiv preprint arXiv:2208.07339},
  year={2022}
}

For 8-bit optimizers or quantization routines, please consider citing the following work:

@article{dettmers2022optimizers,
  title={8-bit Optimizers via Block-wise Quantization},
  author={Dettmers, Tim and Lewis, Mike and Shleifer, Sam and Zettlemoyer, Luke},
  journal={9th International Conference on Learning Representations, ICLR},
  year={2022}
}