Update run_eval.py

run_eval.py

@@ -22,6 +22,7 @@ from logging import getLogger
 from pathlib import Path
 from typing import Dict, List
 
+import torch
 from tqdm import tqdm
 
 from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
@@ -33,12 +34,13 @@ from utils import (
     use_task_specific_params,
 )
 
+from evaluate_gpt import gpt_eval
 
 
 logger = getLogger(__name__)
 
 
-DEFAULT_DEVICE = "cpu"
+DEFAULT_DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
 
 
 def generate_summaries_or_translations(
@@ -206,6 +208,17 @@ def run_generate(
     if scor_path:
         args.score_path = scor_path
 
+    if args.model_name[-3:] == 'gpt':
+        gpt_eval(
+            model_name_path=args.model_name,
+            src_txt=args.input_path,
+            tar_txt=args.reference_path,
+            gen_path=args.save_path,
+            scor_path=args.score_path,
+            batch_size=args.bs
+        )
+        return None
+
     if parsed_args and verbose:
         print(f"parsed the following generate kwargs: {parsed_args}")
     examples = [