Update app.py
Browse files
app.py
CHANGED
|
@@ -1,7 +1,30 @@
|
|
| 1 |
import gradio as gr
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
|
| 3 |
def display_message():
|
| 4 |
-
|
|
|
|
| 5 |
|
| 6 |
iface = gr.Interface(fn=display_message, inputs=None, outputs="text")
|
| 7 |
iface.launch()
|
|
|
|
| 1 |
import gradio as gr
|
| 2 |
+
import torch
|
| 3 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 4 |
+
|
| 5 |
+
def runLLM ():
|
| 6 |
+
model = AutoModelForCausalLM.from_pretrained("cyberagent/open-calm-small", device_map="auto", torch_dtype=torch.float16)
|
| 7 |
+
tokenizer = AutoTokenizer.from_pretrained("cyberagent/open-calm-small")
|
| 8 |
+
|
| 9 |
+
inputs = tokenizer("AIによって私達の暮らしは、", return_tensors="pt").to(model.device)
|
| 10 |
+
|
| 11 |
+
with torch.no_grad():
|
| 12 |
+
tokens = model.generate(
|
| 13 |
+
**inputs,
|
| 14 |
+
max_new_tokens=64,
|
| 15 |
+
do_sample=True,
|
| 16 |
+
temperature=0.7,
|
| 17 |
+
top_p=0.9,
|
| 18 |
+
repetition_penalty=1.05,
|
| 19 |
+
pad_token_id=tokenizer.pad_token_id,
|
| 20 |
+
)
|
| 21 |
+
|
| 22 |
+
output = tokenizer.decode(tokens[0], skip_special_tokens=True)
|
| 23 |
+
return output
|
| 24 |
|
| 25 |
def display_message():
|
| 26 |
+
msg = runLLM()
|
| 27 |
+
return msg
|
| 28 |
|
| 29 |
iface = gr.Interface(fn=display_message, inputs=None, outputs="text")
|
| 30 |
iface.launch()
|