text
stringlengths
4
13
torch
transformers
datasets
tqdm
galore_torch
argparse
accelerate
bitsandbytes
sentencepiece
protobuf
safetensors
tokenizers
peft

SUPPORT ME ON PATREON


Prerequisites:

  1. Python: https://www.python.org/downloads/
  2. Git: https://git-scm.com/downloads

Instructions:

  1. Make sure python and git are installed
  2. Open a command prompt terminal on your local folder
  3. In command prompt run
git lfs install

then

git clone https://huggingface.co/datasets/Rombo-Org/Easy_Galore_8bit_training_With_Native_Windows_Support

then

cd Easy_Galore_8bit_training_With_Native_Windows_Support
  1. Now minimize the command prompt window and open the "Galore_8bit_Version-2.py" file
  2. Edit the paramaters to suite your needs
  3. Open the command prompt window and run
Python Galore_8bit_Version-2.py

The training will now run completely and save your model in the specified folder location.

Downloads last month
14