kwarg change

mlfoundations · Oct 9, 2023 · b51caf9 · b51caf9
1 parent 3f1d66b
commit b51caf9
Showing 1 changed file with 11 additions and 11 deletions.
diff --git a/open_flamingo/eval/evaluate.py b/open_flamingo/eval/evaluate.py
@@ -574,7 +574,7 @@ def main():
             eval_model=eval_model,
             results=results,
             eval_fn=evaluate_vqa,
-            max_generation_length=10,
+            max_new_tokens=10,
         )
 
     if args.eval_imagenet:
@@ -631,8 +631,8 @@ def evaluate_captioning(
     args: argparse.Namespace,
     eval_model: BaseEvalModel,
     seed: int = 42,
-    min_generation_length: int = 0,
-    max_generation_length: int = 20,
+    min_new_tokens: int = 0,
+    max_new_tokens: int = 20,
     num_beams: int = 3,
     length_penalty: float = 0.0,
     num_shots: int = 8,
@@ -645,7 +645,7 @@ def evaluate_captioning(
         args (argparse.Namespace): arguments
         eval_model (BaseEvalModel): model to evaluate
         seed (int, optional): seed for random number generator. Defaults to 42.
-        max_generation_length (int, optional): maximum length of the generated caption. Defaults to 20.
+        max_new_tokens (int, optional): maximum length of the generated caption. Defaults to 20.
         num_beams (int, optional): number of beams to use for beam search. Defaults to 3.
         length_penalty (float, optional): length penalty for beam search. Defaults to -2.0.
         num_shots (int, optional): number of in-context samples to use. Defaults to 8.
@@ -747,8 +747,8 @@ def evaluate_captioning(
         outputs = eval_model.get_outputs(
             batch_images=batch_images,
             batch_text=batch_text,
-            min_generation_length=min_generation_length,
-            max_generation_length=max_generation_length,
+            min_new_tokens=min_new_tokens,
+            max_new_tokens=max_new_tokens,
             num_beams=num_beams,
             length_penalty=length_penalty,
         )
@@ -808,8 +808,8 @@ def evaluate_vqa(
     args: argparse.Namespace,
     eval_model: BaseEvalModel,
     seed: int = 42,
-    min_generation_length: int = 0,
-    max_generation_length: int = 5,
+    min_new_tokens: int = 0,
+    max_new_tokens: int = 5,
     num_beams: int = 3,
     length_penalty: float = 0.0,
     num_shots: int = 8,
@@ -823,7 +823,7 @@ def evaluate_vqa(
         args (argparse.Namespace): arguments
         eval_model (BaseEvalModel): model to evaluate
         seed (int, optional): random seed. Defaults to 42.
-        max_generation_length (int, optional): max generation length. Defaults to 5.
+        max_new_tokens (int, optional): max generation length. Defaults to 5.
         num_beams (int, optional): number of beams to use for beam search. Defaults to 3.
         length_penalty (float, optional): length penalty for beam search. Defaults to -2.0.
         num_shots (int, optional): number of shots to use. Defaults to 8.
@@ -948,8 +948,8 @@ def evaluate_vqa(
         outputs = eval_model.get_outputs(
             batch_images=batch_images,
             batch_text=batch_text,
-            min_generation_length=min_generation_length,
-            max_generation_length=max_generation_length,
+            min_new_tokens=min_new_tokens,
+            max_new_tokens=max_new_tokens,
             num_beams=num_beams,
             length_penalty=length_penalty,
         )