misc

pytorch · Jan 17, 2025 · a80b7e5 · a80b7e5
1 parent 68aee31
commit a80b7e5
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 4 deletions.
diff --git a/recipes/dev/generate_v2.py b/recipes/dev/generate_v2.py
@@ -113,10 +113,10 @@ def log_metrics(self, total_time: int, tokens_per_second: float) -> None:
             f"Time for inference: {total_time:.02f} sec total, {tokens_per_second:.02f} tokens/sec"
         )
         self._logger.info(
-            f"Bandwidth achieved: {model_size * tokens_per_second / 1024 / 1024 / 1024:.02f} GiB/s"
+            f"Bandwidth achieved: {model_size * tokens_per_second / (1024**3):.02f} GiB/s"
         )
         self._logger.info(
-            f"Max memory allocated: {torch.cuda.max_memory_allocated() / 1024 / 1024 / 1024:.02f} GiB"
+            f"Max memory allocated: {torch.cuda.max_memory_allocated() / (1024**3):.02f} GiB"
         )
 
     @torch.inference_mode()

diff --git a/recipes/dev/generate_v2_distributed.py b/recipes/dev/generate_v2_distributed.py
@@ -154,10 +154,10 @@ def log_metrics(self, total_time: int, tokens_per_second: float) -> None:
             f"Time for inference: {total_time:.02f} sec total, {tokens_per_second:.02f} tokens/sec"
         )
         self._logger.info(
-            f"Bandwidth achieved: {model_size * tokens_per_second / 1024 / 1024 / 1024:.02f} GiB/s"
+            f"Bandwidth achieved: {model_size * tokens_per_second / (1024**3):.02f} GiB/s"
         )
         self._logger.info(
-            f"Max memory allocated: {torch.cuda.max_memory_allocated() / 1024 / 1024 / 1024 :.02f} GiB"
+            f"Max memory allocated: {torch.cuda.max_memory_allocated() / (1024**3):.02f} GiB"
         )
 
     @torch.inference_mode()