Update app.py
Browse files
app.py
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
-
import
|
| 2 |
-
gradio as gr
|
| 3 |
from peft import PeftModel, PeftConfig
|
| 4 |
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
|
| 5 |
import torch
|
|
@@ -12,8 +11,7 @@ bnb_config = BitsAndBytesConfig(
|
|
| 12 |
load_in_4bit=True,
|
| 13 |
bnb_4bit_use_double_quant=True,
|
| 14 |
bnb_4bit_quant_type="nf4",
|
| 15 |
-
bnb_4bit_compute_dtype=torch
|
| 16 |
-
.bfloat16
|
| 17 |
)
|
| 18 |
|
| 19 |
# Load models and tokenizer efficiently
|
|
|
|
| 1 |
+
import gradio as gr
|
|
|
|
| 2 |
from peft import PeftModel, PeftConfig
|
| 3 |
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
|
| 4 |
import torch
|
|
|
|
| 11 |
load_in_4bit=True,
|
| 12 |
bnb_4bit_use_double_quant=True,
|
| 13 |
bnb_4bit_quant_type="nf4",
|
| 14 |
+
bnb_4bit_compute_dtype=torch.bfloat16
|
|
|
|
| 15 |
)
|
| 16 |
|
| 17 |
# Load models and tokenizer efficiently
|