github/home-llm
1
1
Fork 0
mirror of https://github.com/acon96/home-llm.git synced 2026-01-10 06:07:58 -05:00
Code Issues Packages Projects Releases Wiki Activity

train script fixes

Browse Source
This commit is contained in:
Alex O'Connell
2025-02-26 08:13:11 -05:00
parent 2712f605a5
commit 136d2bfd3e
3 changed files with 52 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
scripts/convert_and_quantize.sh Normal file → Executable file
View File

@@ -11,8 +11,7 @@ fi
echo "Converting to GGUF..."
if [ ! -f "./models/$MODEL_NAME/$MODEL_NAME.f16.gguf" ]; then
$LLAMA_CPP/convert.py --outfile ./models/$MODEL_NAME/$MODEL_NAME.f16.gguf --outtype f16 ./models/$MODEL_NAME/
# $LLAMA_CPP/convert-hf-to-gguf.py --outfile ./models/$MODEL_NAME/$MODEL_NAME.f16.gguf --outtype f16 ./models/$MODEL_NAME/
$LLAMA_CPP/convert_hf_to_gguf.py --outfile ./models/$MODEL_NAME/$MODEL_NAME.f16.gguf --outtype f16 ./models/$MODEL_NAME/
else
echo "Converted model for already exists. Skipping..."
fi
@@ -23,7 +22,7 @@ for QUANT in "${DESIRED_QUANTS[@]}"
do
QUANT_LOWER=$(echo "$QUANT" | awk '{print tolower($0)}')
if [ ! -f "./models/$MODEL_NAME/$MODEL_NAME.$QUANT_LOWER.gguf" ]; then
$LLAMA_CPP/build/bin/quantize ./models/$MODEL_NAME/$MODEL_NAME.f16.gguf ./models/$MODEL_NAME/$MODEL_NAME.$QUANT_LOWER.gguf $QUANT
$LLAMA_CPP/build/bin/llama-quantize ./models/$MODEL_NAME/$MODEL_NAME.f16.gguf ./models/$MODEL_NAME/$MODEL_NAME.$QUANT_LOWER.gguf $QUANT
else
echo "Quantized model for '$QUANT' already exists. Skipping..."
fi