suvadityamuk commited on
Commit
f033a76
·
1 Parent(s): 7998f0c

Signed-off-by: Suvaditya Mukherjee <[email protected]>

Files changed (1) hide show
  1. app.py +7 -7
app.py CHANGED
@@ -194,14 +194,14 @@ if __name__ == "__main__":
194
  model_name,
195
  torch_dtype=torch.bfloat16,
196
  device_map="auto",
197
- # quantization_config=QuantoConfig(
198
- # weights="int4",
199
- # )
200
- quantization_config = BitsAndBytesConfig(
201
- load_in_8bit=True,
202
- # bnb_4bit_compute_dtype=torch.float16,
203
- # bnb_4bit_quant_type="nf4"
204
  )
 
 
 
 
 
205
  )
206
 
207
  tokenizer = AutoTokenizer.from_pretrained(model_name)
 
194
  model_name,
195
  torch_dtype=torch.bfloat16,
196
  device_map="auto",
197
+ quantization_config=QuantoConfig(
198
+ weights="int4",
 
 
 
 
 
199
  )
200
+ # quantization_config = BitsAndBytesConfig(
201
+ # load_in_8bit=True,
202
+ # # bnb_4bit_compute_dtype=torch.float16,
203
+ # # bnb_4bit_quant_type="nf4"
204
+ # )
205
  )
206
 
207
  tokenizer = AutoTokenizer.from_pretrained(model_name)