fix: use 0-indexing consistently (shard 0-15)

2026-07-03 00:14:57 -04:00
parent 262b1181b7
commit 50623cdcd0
1 changed files with 1 additions and 1 deletions
--- a/quantize_streaming.py
+++ b/quantize_streaming.py
@@ -47,7 +47,7 @@ def streaming_quantize(model_path: str, output_path: str):
        if (output_path / shard_name).exists():
            return
        
-        print(f"[{idx+1}/{len(shards)}] {Path(shard_file).name} (GPU {gpu_id})")
+        print(f"[{idx}/{len(shards)}] {Path(shard_file).name} (GPU {gpu_id})")
        
        # Load to assigned GPU
        state_dict = load_file(shard_file, device=f"cuda:{gpu_id}")