Spaces:
Sleeping
Sleeping
Update main.py
Browse files
main.py
CHANGED
@@ -1,14 +1,20 @@
|
|
1 |
#Imporing required libraries
|
2 |
from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
|
3 |
import gradio as gr
|
|
|
4 |
|
5 |
# Defining the pipeline and the model
|
6 |
|
|
|
|
|
7 |
tokenizer = AutoTokenizer.from_pretrained("huggyllama/llama-7b")
|
8 |
-
model = AutoModelForCausalLM.from_pretrained("huggyllama/llama-7b")
|
|
|
|
|
9 |
|
10 |
print("***")
|
11 |
print("Loaded tokenizer and model")
|
|
|
12 |
print("***")
|
13 |
|
14 |
pipe_flan = pipeline("text-generation", model=model, tokenizer=tokenizer)
|
|
|
1 |
#Imporing required libraries
|
2 |
from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
|
3 |
import gradio as gr
|
4 |
+
import torch
|
5 |
|
6 |
# Defining the pipeline and the model
|
7 |
|
8 |
+
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
9 |
+
|
10 |
tokenizer = AutoTokenizer.from_pretrained("huggyllama/llama-7b")
|
11 |
+
model = AutoModelForCausalLM.from_pretrained("huggyllama/llama-7b", device_map="auto", torch_dtype=torch.float16)
|
12 |
+
|
13 |
+
#model = model.to(device)
|
14 |
|
15 |
print("***")
|
16 |
print("Loaded tokenizer and model")
|
17 |
+
print(device)
|
18 |
print("***")
|
19 |
|
20 |
pipe_flan = pipeline("text-generation", model=model, tokenizer=tokenizer)
|