File size: 689 Bytes
3bdb9f0
2d71f41
 
 
abd754d
3bdb9f0
a3beba9
9da3581
a3beba9
 
dfc3b85
d942439
7469f5d
 
 
 
 
 
 
 
 
 
be6172f
06ffcdd
d942439
4d5ed6a
d942439
3bdb9f0
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
import gradio
from os import system

system("pip3 install torch")
system("pip3 install transformers")

from transformers import AutoTokenizer,AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("apple/OpenELM-270M")
model = openelm_270m = AutoModelForCausalLM.from_pretrained("apple/OpenELM-270M", trust_remote_code=True)

def work(inp_text):
    out = tokenizer.encode(inp_text,return_tensors="pt")

    out = model.generate(
        out,
        max_new_tokens=20,
        do_sample=True,
        temperature=0.3,
    )

    out = tokenizer.decode(out[0])

    return str(out)

demo = gradio.Interface(
    fn=work,
    inputs=["text"],
    outputs=["text"],
)

demo.launch()