Spaces:
Sleeping
Sleeping
Added clean_text function to remove filler words and extra spaces
Browse files- app.py +21 -6
- requirements.txt +2 -1
app.py
CHANGED
@@ -2,6 +2,7 @@ from fastapi import FastAPI, HTTPException
|
|
2 |
from pydantic import BaseModel
|
3 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
4 |
import torch
|
|
|
5 |
|
6 |
app = FastAPI()
|
7 |
|
@@ -11,6 +12,12 @@ model = AutoModelForSequenceClassification.from_pretrained("cardiffnlp/twitter-r
|
|
11 |
class TextRequest(BaseModel):
|
12 |
text: str
|
13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
@app.get("/")
|
15 |
def home():
|
16 |
return {"message": "Speak your mind emotion API is running"}
|
@@ -18,20 +25,28 @@ def home():
|
|
18 |
@app.post("/classify-emotion")
|
19 |
async def classify_emotion(request: TextRequest):
|
20 |
try:
|
21 |
-
text = request.text
|
|
|
|
|
|
|
|
|
22 |
|
23 |
-
inputs = tokenizer(
|
24 |
|
25 |
with torch.no_grad():
|
26 |
outputs = model(**inputs)
|
27 |
|
28 |
-
|
29 |
logits = outputs.logits
|
30 |
predicted_class_id = torch.argmax(logits, dim=-1).item()
|
31 |
-
predicted_emotion = model.config.id2label[predicted_class_id]
|
32 |
|
33 |
-
return {
|
|
|
|
|
|
|
|
|
34 |
|
35 |
except Exception as e:
|
36 |
-
raise HTTPException(status_code=500, detail=str(e))
|
|
|
37 |
|
|
|
2 |
from pydantic import BaseModel
|
3 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
4 |
import torch
|
5 |
+
import re
|
6 |
|
7 |
app = FastAPI()
|
8 |
|
|
|
12 |
class TextRequest(BaseModel):
|
13 |
text: str
|
14 |
|
15 |
+
def clean_text(text: str) -> str:
|
16 |
+
fillers = ["um", "uh", "like", "you know", "I mean", "sort of", "kind of", "hmm", "uhh"]
|
17 |
+
text = re.sub(r'\b(?:' + '|'.join(fillers) + r')\b', '', text, flags=re.IGNORECASE)
|
18 |
+
text = re.sub(r'\s+', ' ', text).strip()
|
19 |
+
return text
|
20 |
+
|
21 |
@app.get("/")
|
22 |
def home():
|
23 |
return {"message": "Speak your mind emotion API is running"}
|
|
|
25 |
@app.post("/classify-emotion")
|
26 |
async def classify_emotion(request: TextRequest):
|
27 |
try:
|
28 |
+
text = request.text.strip()
|
29 |
+
|
30 |
+
if not text:
|
31 |
+
raise HTTPException(status_code=400, detail="Text cannot be empty")
|
32 |
+
cleaned_text = clean_text(text)
|
33 |
|
34 |
+
inputs = tokenizer(cleaned_text, return_tensors="pt", truncation=True, padding=True, max_length=512)
|
35 |
|
36 |
with torch.no_grad():
|
37 |
outputs = model(**inputs)
|
38 |
|
|
|
39 |
logits = outputs.logits
|
40 |
predicted_class_id = torch.argmax(logits, dim=-1).item()
|
41 |
+
predicted_emotion = model.config.id2label[predicted_class_id]
|
42 |
|
43 |
+
return {
|
44 |
+
"original_text": text,
|
45 |
+
"cleaned_text": cleaned_text,
|
46 |
+
"predicted_emotion": predicted_emotion
|
47 |
+
}
|
48 |
|
49 |
except Exception as e:
|
50 |
+
raise HTTPException(status_code=500, detail=f"Error processing text: {str(e)}")
|
51 |
+
|
52 |
|
requirements.txt
CHANGED
@@ -3,4 +3,5 @@ uvicorn
|
|
3 |
transformers
|
4 |
torch
|
5 |
httpx
|
6 |
-
pytest
|
|
|
|
3 |
transformers
|
4 |
torch
|
5 |
httpx
|
6 |
+
pytest
|
7 |
+
pydantic
|