haraldwolff/llama.cpp
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

docker : add finetune option (#4211)

Browse source
This commit is contained in:
Juraj Bednar 2023-11-30 22:46:01 +01:00 committed by GitHub
parent bde629bb53
commit 3bd2c7ce1b
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 4 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
.devops/tools.sh
View file

@ -13,6 +13,8 @@ elif [[ "$arg1" == '--quantize' || "$arg1" == '-q' ]]; then
./quantize "$@"
elif [[ "$arg1" == '--run' || "$arg1" == '-r' ]]; then
./main "$@"
elif [[ "$arg1" == '--finetune' || "$arg1" == '-f' ]]; then
./finetune "$@"
elif [[ "$arg1" == '--all-in-one' || "$arg1" == '-a' ]]; then
echo "Converting PTH to GGML..."
for i in `ls $1/$2/ggml-model-f16.bin*`; do
@ -34,6 +36,8 @@ else
echo " ex: --outtype f16 \"/models/7B/\" "
echo " --quantize (-q): Optimize with quantization process ggml"
echo " ex: \"/models/7B/ggml-model-f16.bin\" \"/models/7B/ggml-model-q4_0.bin\" 2"
echo " --finetune (-f): Run finetune command to create a lora finetune of the model"
echo " See documentation for finetune for command-line parameters"
echo " --all-in-one (-a): Execute --convert & --quantize"
echo " ex: \"/models/\" 7B"
echo " --server (-s): Run a model on the server"