Use array append

app.py

@@ -221,20 +221,24 @@ def quantize_model(
     quantize_cmd = ["llama-quantize"]
 
     if quant_embedding:
-        quantize_cmd += ["--token-embedding-type", embedding_tensor_method]
+        quantize_cmd.append("--token-embedding-type")
+        quantize_cmd.append(embedding_tensor_method)
     if quant_output:
-        quantize_cmd += ["--output-tensor-type", output_tensor_method]
+        quantize_cmd.append("--output-tensor-type")
+        quantize_cmd.append(output_tensor_method)
 
     if use_imatrix:
-        quantize_cmd += ["--imatrix", imatrix_path]
+        quantize_cmd.append("--imatrix")
+        quantize_cmd.append(imatrix_path)
 
     quantized_gguf = str(Path(outdir)/gguf_name)
-    quantize_cmd += [fp16, quantized_gguf]
+    quantize_cmd.append(fp16)
+    quantize_cmd.append(quantized_gguf)
 
     if use_imatrix:
-        quantize_cmd += [imatrix_q_method]
+        quantize_cmd.append(imatrix_q_method)
     else:
-        quantize_cmd += [q_method]
+        quantize_cmd.append(q_method)
 
     print(f"Quantizing model with {quantize_cmd}")
     result = subprocess.run(quantize_cmd, shell=False, capture_output=True)