|
import os |
|
import subprocess |
|
import streamlit as st |
|
from huggingface_hub import snapshot_download, login |
|
|
|
if "quantized_model_path" not in st.session_state: |
|
st.session_state.quantized_model_path = None |
|
if "upload_to_hf" not in st.session_state: |
|
st.session_state.upload_to_hf = False |
|
|
|
def check_directory_path(directory_name: str) -> str: |
|
if os.path.exists(directory_name): |
|
path = os.path.abspath(directory_name) |
|
return str(path) |
|
|
|
|
|
QUANT_TYPES = [ |
|
"Q2_K", "Q3_K_M", "Q3_K_S", "Q4_K_M", "Q4_K_S", |
|
"Q5_K_M", "Q5_K_S", "Q6_K" |
|
] |
|
|
|
model_dir_path = check_directory_path("/app/llama.cpp") |
|
|
|
def download_model(hf_model_name, output_dir="/tmp/models"): |
|
""" |
|
Downloads a Hugging Face model and saves it locally. |
|
""" |
|
st.write(f"π₯ Downloading `{hf_model_name}` from Hugging Face...") |
|
os.makedirs(output_dir, exist_ok=True) |
|
snapshot_download(repo_id=hf_model_name, local_dir=output_dir, local_dir_use_symlinks=False) |
|
st.success("β
Model downloaded successfully!") |
|
|
|
def convert_to_gguf(model_dir, output_file): |
|
""" |
|
Converts a Hugging Face model to GGUF format. |
|
""" |
|
st.write(f"π Converting `{model_dir}` to GGUF format...") |
|
os.makedirs(os.path.dirname(output_file), exist_ok=True) |
|
cmd = [ |
|
"python3", "/app/llama.cpp/convert_hf_to_gguf.py", model_dir, |
|
"--outtype", "f16", "--outfile", output_file |
|
] |
|
process = subprocess.run(cmd, text=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE) |
|
if process.returncode == 0: |
|
st.success(f"β
Conversion complete: `{output_file}`") |
|
else: |
|
st.error(f"β Conversion failed: {process.stderr}") |
|
|
|
def quantize_llama(model_path, quantized_output_path, quant_type): |
|
""" |
|
Quantizes a GGUF model. |
|
""" |
|
st.write(f"β‘ Quantizing `{model_path}` with `{quant_type}` precision...") |
|
os.makedirs(os.path.dirname(quantized_output_path), exist_ok=True) |
|
quantize_path = "/app/llama.cpp/build/bin/llama-quantize" |
|
|
|
cmd = [ |
|
"/app/llama.cpp/build/bin/llama-quantize", |
|
model_path, |
|
quantized_output_path, |
|
quant_type |
|
] |
|
|
|
process = subprocess.run(cmd, text=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE) |
|
|
|
if process.returncode == 0: |
|
st.success(f"β
Quantized model saved at `{quantized_output_path}`") |
|
else: |
|
st.error(f"β Quantization failed: {process.stderr}") |
|
|
|
def automate_llama_quantization(hf_model_name, quant_type): |
|
""" |
|
Orchestrates the entire quantization process. |
|
""" |
|
output_dir = "/tmp/models" |
|
gguf_file = os.path.join(output_dir, f"{hf_model_name.replace('/', '_')}.gguf") |
|
quantized_file = gguf_file.replace(".gguf", f"-{quant_type}.gguf") |
|
|
|
progress_bar = st.progress(0) |
|
|
|
|
|
st.write("### Step 1: Downloading Model") |
|
download_model(hf_model_name, output_dir) |
|
progress_bar.progress(33) |
|
|
|
|
|
st.write("### Step 2: Converting Model to GGUF Format") |
|
convert_to_gguf(output_dir, gguf_file) |
|
progress_bar.progress(66) |
|
|
|
|
|
st.write("### Step 3: Quantizing Model") |
|
quantize_llama(gguf_file, quantized_file, quant_type.lower()) |
|
progress_bar.progress(100) |
|
|
|
st.success(f"π All steps completed! Quantized model available at: `{quantized_file}`") |
|
return quantized_file |
|
|
|
def upload_to_huggingface(file_path, repo_id, token): |
|
""" |
|
Uploads a file to Hugging Face Hub. |
|
""" |
|
try: |
|
|
|
login(token=token) |
|
|
|
|
|
api = HfApi() |
|
|
|
|
|
api.create_repo(repo_id, exist_ok=True, repo_type="model") |
|
|
|
|
|
api.upload_file( |
|
path_or_fileobj=file_path, |
|
path_in_repo=os.path.basename(file_path), |
|
repo_id=repo_id, |
|
) |
|
st.success(f"β
File uploaded to Hugging Face: {repo_id}") |
|
except Exception as e: |
|
st.error(f"β Failed to upload file: {e}") |
|
|
|
st.title("π¦ LLaMA Model Quantization (llama.cpp)") |
|
|
|
hf_model_name = st.text_input("Enter Hugging Face Model Name", "Qwen/Qwen2.5-1.5B") |
|
quant_type = st.selectbox("Select Quantization Type", QUANT_TYPES) |
|
start_button = st.button("π Start Quantization") |
|
|
|
if start_button: |
|
with st.spinner("Processing..."): |
|
st.session_state.quantized_model_path = automate_llama_quantization(hf_model_name, quant_type) |
|
|
|
if st.session_state.quantized_model_path: |
|
with open(st.session_state.quantized_model_path, "rb") as f: |
|
st.download_button("β¬οΈ Download Quantized Model", f, file_name=os.path.basename(st.session_state.quantized_model_path)) |
|
|
|
|
|
st.session_state.upload_to_hf = st.checkbox("Upload to Hugging Face", value=st.session_state.upload_to_hf) |
|
|
|
if st.session_state.upload_to_hf: |
|
st.write("### Upload to Hugging Face") |
|
repo_id = st.text_input("Enter Hugging Face Repository ID (e.g., 'username/repo-name')") |
|
hf_token = st.text_input("Enter Hugging Face Token", type="password") |
|
|
|
if st.button("π€ Upload to Hugging Face"): |
|
if repo_id and hf_token: |
|
with st.spinner("Uploading..."): |
|
upload_to_huggingface(st.session_state.quantized_model_path, repo_id, hf_token) |
|
else: |
|
st.warning("Please provide a valid repository ID and Hugging Face token.") |