Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -1,16 +1,18 @@
|
|
1 |
import gradio as gr
|
2 |
-
from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer
|
3 |
import torch
|
4 |
|
5 |
model_id = "thrishala/mental_health_chatbot"
|
6 |
|
7 |
try:
|
|
|
|
|
|
|
8 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
9 |
model = AutoModelForCausalLM.from_pretrained(
|
10 |
model_id,
|
11 |
-
|
12 |
device_map="auto", #Use GPU if available
|
13 |
-
torch_dtype=torch.float16 #Use float 16 for additional memory reduction
|
14 |
)
|
15 |
pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
|
16 |
|
|
|
1 |
import gradio as gr
|
2 |
+
from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
|
3 |
import torch
|
4 |
|
5 |
model_id = "thrishala/mental_health_chatbot"
|
6 |
|
7 |
try:
|
8 |
+
quantization_config = BitsAndBytesConfig(
|
9 |
+
load_in_4bit=True, # we going to 4 babey
|
10 |
+
)
|
11 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
12 |
model = AutoModelForCausalLM.from_pretrained(
|
13 |
model_id,
|
14 |
+
quantization_config=quantization_config,
|
15 |
device_map="auto", #Use GPU if available
|
|
|
16 |
)
|
17 |
pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
|
18 |
|