Spaces:
Paused
Paused
Update main.py
Browse files
main.py
CHANGED
@@ -25,15 +25,10 @@ app.add_middleware(
|
|
25 |
allow_headers=["*"],
|
26 |
)
|
27 |
|
28 |
-
def chunk_text(text, chunk_size=512):
|
29 |
-
return [text[i:i + chunk_size] for i in range(0, len(text), chunk_size)]
|
30 |
|
31 |
@app.post("/get_embeding")
|
32 |
-
async def get_embeding(
|
33 |
-
|
34 |
-
all_embeddings = []
|
35 |
-
for chunk in chunks:
|
36 |
-
|
37 |
# Tokenize the input text
|
38 |
inputs = tokenizer(chunk, return_tensors="pt")
|
39 |
|
@@ -47,9 +42,9 @@ async def get_embeding(text):
|
|
47 |
# Optionally, you can average the token embeddings to get a single vector for the sentence
|
48 |
sentence_embedding = torch.mean(embeddings, dim=1)
|
49 |
|
50 |
-
print(sentence_embedding)
|
51 |
-
|
52 |
-
|
53 |
|
54 |
|
55 |
|
|
|
25 |
allow_headers=["*"],
|
26 |
)
|
27 |
|
|
|
|
|
28 |
|
29 |
@app.post("/get_embeding")
|
30 |
+
async def get_embeding(chunk):
|
31 |
+
|
|
|
|
|
|
|
32 |
# Tokenize the input text
|
33 |
inputs = tokenizer(chunk, return_tensors="pt")
|
34 |
|
|
|
42 |
# Optionally, you can average the token embeddings to get a single vector for the sentence
|
43 |
sentence_embedding = torch.mean(embeddings, dim=1)
|
44 |
|
45 |
+
#print(sentence_embedding)
|
46 |
+
return sentence_embedding.tolist()
|
47 |
+
|
48 |
|
49 |
|
50 |
|