Update app.py

app.py

@@ -11,10 +11,8 @@ from gradio_huggingfacehub_search import HuggingfaceHubSearch
 from apscheduler.schedulers.background import BackgroundScheduler
 from datetime import datetime
 import numpy as np
-import shutil
-
-
 HF_TOKEN = os.environ.get("HF_TOKEN")
+
 os.environ["GRADIO_ANALYTICS_ENABLED"] = "False"
 CONVERSION_SCRIPT = "./llama.cpp/convert_hf_to_gguf.py"
 
@@ -227,19 +225,21 @@ def process_model(model_id, q_method, use_imatrix, imatrix_q_method, private_rep
           fp16 = str(Path(outdir)/f"{model_name}.fp16.gguf")
 
           with tempfile.TemporaryDirectory(dir=downloads_dir) as tmpdir:
+              print("Downloading")
               local_dir = Path(tmpdir)/model_name
-              print(datetime.now().strftime("%Y-%m-%d %H:%M:%S") + " Start download")
               api.snapshot_download(repo_id=model_id, local_dir=local_dir, local_dir_use_symlinks=False, allow_patterns=dl_pattern)
-
+              
               config_dir = local_dir/"config.json"
               adapter_config_dir = local_dir/"adapter_config.json"
               if os.path.exists(adapter_config_dir) and not os.path.exists(config_dir):
                   raise Exception("adapter_config.json is present. If converting LoRA, use GGUF-my-lora.")
-              print(datetime.now().strftime("%Y-%m-%d %H:%M:%S")+" Download finished and start converting to fp16")
+
+              print("Download successfully")
               result = subprocess.run(["python", CONVERSION_SCRIPT, local_dir, "--outtype", "f16", "--outfile", fp16], shell=False, capture_output=True)
+              print("Converted to f16")
               if result.returncode != 0:
                   raise Exception(f"Error converting to fp16: {result.stderr.decode()}")
-          shutil.rmtree(downloads_dir)
+
           imatrix_path = Path(outdir)/"imatrix.dat"
           if use_imatrix:
               train_data_path = train_data_file.name if train_data_file else "llama.cpp/groups_merged.txt"
@@ -249,9 +249,10 @@ def process_model(model_id, q_method, use_imatrix, imatrix_q_method, private_rep
 
           quant_methods = [imatrix_q_method] if use_imatrix else (q_method if isinstance(q_method, list) else [q_method])
           suffix = "imat" if use_imatrix else None
-          print(datetime.now().strftime("%Y-%m-%d %H:%M:%S")+" Converting finished, start to quantize")
+
           gguf_files = []
           for method in quant_methods:
+              print("Begin quantize")
               name = f"{model_name.lower()}-{method.lower()}-{suffix}.gguf" if suffix else f"{model_name.lower()}-{method.lower()}.gguf"
               path = str(Path(outdir)/name)
               quant_cmd = ["./llama.cpp/llama-quantize", "--imatrix", imatrix_path, fp16, path, method] if use_imatrix else ["./llama.cpp/llama-quantize", fp16, path, method]
@@ -261,6 +262,7 @@ def process_model(model_id, q_method, use_imatrix, imatrix_q_method, private_rep
               size = os.path.getsize(path)/1024/1024/1024
               gguf_files.append((name, path, size, method))
 
+          print("Quantize successfully!")
           suffix_for_repo = f"{imatrix_q_method}-imat" if use_imatrix else "-".join(quant_methods)
           repo_id = f"{repo_namespace}/{model_name}-{suffix_for_repo}-GGUF"
           new_repo_url = api.create_repo(repo_id=repo_id, exist_ok=True, private=private_repo)