Spaces:
Sleeping
Sleeping
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from transformers import AutoTokenizer, AutoModel
|
3 |
+
import torch
|
4 |
+
|
5 |
+
tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')
|
6 |
+
model = AutoModel.from_pretrained('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')
|
7 |
+
|
8 |
+
def mean_pooling(model_output, attention_mask):
|
9 |
+
token_embeddings = model_output[0]
|
10 |
+
input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
|
11 |
+
return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
|
12 |
+
|
13 |
+
def encode_sentences(sentences):
|
14 |
+
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
|
15 |
+
with torch.no_grad():
|
16 |
+
model_output = model(**encoded_input)
|
17 |
+
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
|
18 |
+
return sentence_embeddings.tolist()
|
19 |
+
|
20 |
+
demo = gr.Interface(fn=encode_sentences,
|
21 |
+
inputs="textbox",
|
22 |
+
outputs="text")
|
23 |
+
|
24 |
+
if __name__ == "__main__":
|
25 |
+
demo.launch()
|