Spaces:
Build error
Build error
Mandar Patil
commited on
Commit
•
2c050f2
1
Parent(s):
06b224c
Add application file
Browse files
app.py
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
|
2 |
+
import torch
|
3 |
+
import gradio as gr
|
4 |
+
from transformers import BlenderbotTokenizer
|
5 |
+
from transformers import BlenderbotTokenizer, BlenderbotForConditionalGeneration, BlenderbotConfig
|
6 |
+
from transformers import BlenderbotTokenizerFast
|
7 |
+
import contextlib
|
8 |
+
|
9 |
+
#tokenizer = AutoTokenizer.from_pretrained("facebook/blenderbot-400M-distill")
|
10 |
+
#model = AutoModelForSeq2SeqLM.from_pretrained("facebook/blenderbot-400M-distill")
|
11 |
+
#tokenizer = AutoTokenizer.from_pretrained("facebook/blenderbot-3B")
|
12 |
+
mname = "facebook/blenderbot-400M-distill"
|
13 |
+
#configuration = BlenderbotConfig.from_pretrained(mname)
|
14 |
+
tokenizer = BlenderbotTokenizerFast.from_pretrained(mname)
|
15 |
+
model = BlenderbotForConditionalGeneration.from_pretrained(mname)
|
16 |
+
#tokenizer = BlenderbotTokenizer.from_pretrained(mname)
|
17 |
+
#-----------new chat-----------
|
18 |
+
print(mname + 'model loaded')
|
19 |
+
def predict(input,history=[]):
|
20 |
+
|
21 |
+
history.append(input)
|
22 |
+
|
23 |
+
listToStr= '</s> <s>'.join([str(elem)for elem in history[len(history)-3:]])
|
24 |
+
#print('listToStr -->',str(listToStr))
|
25 |
+
input_ids = tokenizer([(listToStr)], return_tensors="pt",max_length=512,truncation=True)
|
26 |
+
next_reply_ids = model.generate(**input_ids,max_length=512, pad_token_id=tokenizer.eos_token_id)
|
27 |
+
response = tokenizer.batch_decode(next_reply_ids, skip_special_tokens=True)[0]
|
28 |
+
history.append(response)
|
29 |
+
response = [(history[i], history[i+1]) for i in range(0, len(history)-1, 2)] # convert to tuples of list
|
30 |
+
return response, history
|
31 |
+
|
32 |
+
demo = gr.Interface(fn=predict, inputs=["text",'state'], outputs=["chatbot",'state'])
|
33 |
+
demo.launch(share=True)
|