commit

training/generate.py

@@ -33,13 +33,13 @@ def load_model_tokenizer_for_generate(
     Returns:
         Tuple[PreTrainedModel, PreTrainedTokenizer]: model and tokenizer
     """
-    # tokenizer = AutoTokenizer.from_pretrained(pretrained_model_name_or_path, padding_side="left")#, cache_dir="/media/siiva/DataStore/LLMs/cache/dollyV2")
+    tokenizer = AutoTokenizer.from_pretrained(pretrained_model_name_or_path, padding_side="left")#, cache_dir="/media/siiva/DataStore/LLMs/cache/dollyV2")
     # model = AutoModelForCausalLM.from_pretrained(
     #     pretrained_model_name_or_path, device_map="auto", trust_remote_code=True)#, cache_dir="/media/siiva/DataStore/LLMs/cache/dollyV2"
     #)
     model = AutoModelForCausalLM.from_pretrained(pretrained_model_name_or_path, model_type='dolly-v2')
 
-    tokenizer = AutoTokenizer.from_pretrained('gpt2')
+    # tokenizer = AutoTokenizer.from_pretrained('gpt2')
 
 
     return model, tokenizer