Merge pull request #1409 from brian41005/master

Evaluation result.txt path changing #1286

examples/run_glue.py

@@ -249,7 +249,7 @@ def evaluate(args, model, tokenizer, prefix=""):
         result = compute_metrics(eval_task, preds, out_label_ids)
         results.update(result)
 
-        output_eval_file = os.path.join(eval_output_dir, "eval_results.txt")
+        output_eval_file = os.path.join(eval_output_dir, prefix, "eval_results.txt")
         with open(output_eval_file, "w") as writer:
             logger.info("***** Eval results {} *****".format(prefix))
             for key in sorted(result.keys()):
@@ -490,9 +490,11 @@ def main():
         logger.info("Evaluate the following checkpoints: %s", checkpoints)
         for checkpoint in checkpoints:
             global_step = checkpoint.split('-')[-1] if len(checkpoints) > 1 else ""
+            prefix = checkpoint.split('/')[-1] if checkpoint.find('checkpoint') != -1 else ""
+            
             model = model_class.from_pretrained(checkpoint)
             model.to(args.device)
-            result = evaluate(args, model, tokenizer, prefix=global_step)
+            result = evaluate(args, model, tokenizer, prefix=prefix)
             result = dict((k + '_{}'.format(global_step), v) for k, v in result.items())
             results.update(result)
 

examples/run_lm_finetuning.py

@@ -282,7 +282,7 @@ def evaluate(args, model, tokenizer, prefix=""):
         "perplexity": perplexity
     }
 
-    output_eval_file = os.path.join(eval_output_dir, "eval_results.txt")
+    output_eval_file = os.path.join(eval_output_dir, prefix, "eval_results.txt")
     with open(output_eval_file, "w") as writer:
         logger.info("***** Eval results {} *****".format(prefix))
         for key in sorted(result.keys()):
@@ -484,9 +484,11 @@ def main():
         logger.info("Evaluate the following checkpoints: %s", checkpoints)
         for checkpoint in checkpoints:
             global_step = checkpoint.split('-')[-1] if len(checkpoints) > 1 else ""
+            prefix = checkpoint.split('/')[-1] if checkpoint.find('checkpoint') != -1 else ""
+            
             model = model_class.from_pretrained(checkpoint)
             model.to(args.device)
-            result = evaluate(args, model, tokenizer, prefix=global_step)
+            result = evaluate(args, model, tokenizer, prefix=prefix)
             result = dict((k + '_{}'.format(global_step), v) for k, v in result.items())
             results.update(result)
 

examples/run_multiple_choice.py

@@ -512,9 +512,11 @@ def main():
         logger.info("Evaluate the following checkpoints: %s", checkpoints)
         for checkpoint in checkpoints:
             global_step = checkpoint.split('-')[-1] if len(checkpoints) > 1 else ""
+            prefix = checkpoint.split('/')[-1] if checkpoint.find('checkpoint') != -1 else ""
+            
             model = model_class.from_pretrained(checkpoint)
             model.to(args.device)
-            result = evaluate(args, model, tokenizer, prefix=global_step)
+            result = evaluate(args, model, tokenizer, prefix=prefix)
             result = dict((k + '_{}'.format(global_step), v) for k, v in result.items())
             results.update(result)
 
@@ -528,9 +530,11 @@ def main():
         logger.info("Evaluate the following checkpoints: %s", checkpoints)
         for checkpoint in checkpoints:
             global_step = checkpoint.split('-')[-1] if len(checkpoints) > 1 else ""
+            prefix = checkpoint.split('/')[-1] if checkpoint.find('checkpoint') != -1 else ""
+            
             model = model_class.from_pretrained(checkpoint)
             model.to(args.device)
-            result = evaluate(args, model, tokenizer, prefix=global_step, test=True)
+            result = evaluate(args, model, tokenizer, prefix=prefix, test=True)
             result = dict((k + '_{}'.format(global_step), v) for k, v in result.items())
             results.update(result)
     if best_steps: