Update app.py

app.py

@@ -6,7 +6,7 @@ import torch
 # from peft import PeftModel, PeftConfig
 
 
-tokenizer = AutoTokenizer.from_pretrained("FlawedLLM/Bhashini")
+# tokenizer = AutoTokenizer.from_pretrained("FlawedLLM/Bhashini")
 # quantization_config = BitsAndBytesConfig(
 #         load_in_4bit=True,
 #         bnb_4bit_use_double_quant=True,
@@ -34,8 +34,15 @@ tokenizer = AutoTokenizer.from_pretrained("FlawedLLM/Bhashini")
 
 
 
-model = AutoModel.from_pretrained("FlawedLLM/Bhashini", load_in_4bit=True, device_map='auto')
-
+# model = AutoModel.from_pretrained("FlawedLLM/Bhashini", load_in_4bit=True, device_map='auto')
+    # I highly do NOT suggest - use Unsloth if possible
+from peft import AutoPeftModelForCausalLM
+from transformers import AutoTokenizer
+model = AutoPeftModelForCausalLM.from_pretrained(
+        "FlawedLLM/Bhashini", # YOUR MODEL YOU USED FOR TRAINING
+        load_in_4bit = load_in_4bit,
+    )
+tokenizer = AutoTokenizer.from_pretrained("FlawedLLM/Bhashini")
 
 @spaces.GPU(duration=300)
 def chunk_it(input_command):