ranamhamoud commited on
Commit
89044b5
·
verified ·
1 Parent(s): 4d5d8af

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +12 -12
app.py CHANGED
@@ -26,18 +26,18 @@ if not torch.cuda.is_available():
26
 
27
  if torch.cuda.is_available():
28
 
29
- # Model and Tokenizer Configuration
30
- model_id = "meta-llama/Llama-2-7b-chat-hf"
31
- bnb_config = BitsAndBytesConfig(
32
- load_in_4bit=True,
33
- bnb_4bit_use_double_quant=False,
34
- bnb_4bit_quant_type="nf4",
35
- bnb_4bit_compute_dtype=torch.bfloat16
36
- )
37
- base_model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", quantization_config=bnb_config)
38
- model = PeftModel.from_pretrained(base_model, "ranamhamoud/storytell")
39
- tokenizer = AutoTokenizer.from_pretrained(model_id)
40
- tokenizer.pad_token = tokenizer.eos_token
41
 
42
  # # MongoDB Connection
43
  # PASSWORD = os.environ.get("MONGO_PASS")
 
26
 
27
  if torch.cuda.is_available():
28
 
29
+ # Model and Tokenizer Configuration
30
+ model_id = "meta-llama/Llama-2-7b-chat-hf"
31
+ bnb_config = BitsAndBytesConfig(
32
+ load_in_4bit=True,
33
+ bnb_4bit_use_double_quant=False,
34
+ bnb_4bit_quant_type="nf4",
35
+ bnb_4bit_compute_dtype=torch.bfloat16
36
+ )
37
+ base_model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", quantization_config=bnb_config)
38
+ model = PeftModel.from_pretrained(base_model, "ranamhamoud/storytell")
39
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
40
+ tokenizer.pad_token = tokenizer.eos_token
41
 
42
  # # MongoDB Connection
43
  # PASSWORD = os.environ.get("MONGO_PASS")