Rename app.py to app_24_4_2024.py

app.py → app_24_4_2024.py

@@ -3,6 +3,7 @@ from transformers import AutoProcessor, Idefics2ForConditionalGeneration, AutoMo
 import subprocess
 import torch
 from peft import LoraConfig
+from huggingface_hub import InferenceApi
 # from transformers import BitsAndBytesConfig
 
 subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
@@ -17,7 +18,7 @@ processor = AutoProcessor.from_pretrained("HuggingFaceM4/idefics2-8b", do_image_
 #         bnb_4bit_compute_dtype=torch.float16
 #     ) 
 
-# model = AutoModelForPreTraining.from_pretrained("HuggingFaceM4/idefics2-8b",quantization_config=bnb_config)
+
 model = Idefics2ForConditionalGeneration.from_pretrained("HuggingFaceM4/idefics2-8b",load_in_4bit=True)
 
 # if USE_QLORA or USE_LORA: