Spaces:
Sleeping
Sleeping
Commit
·
e3bd131
1
Parent(s):
e548c5b
Add application file
Browse files
app.py
CHANGED
@@ -1,4 +1,59 @@
|
|
1 |
import streamlit as st
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
|
3 |
-
x = st.slider('Select a value')
|
4 |
-
st.write(x, 'squared is', x * x)
|
|
|
1 |
import streamlit as st
|
2 |
+
import torch
|
3 |
+
from transformers import BertTokenizer, BertForSequenceClassification
|
4 |
+
|
5 |
+
# Load the tokenizer and model
|
6 |
+
tokenizer = BertTokenizer.from_pretrained('caesarCITREA/crocus-bert-medical-department-classification')
|
7 |
+
model = BertForSequenceClassification.from_pretrained('caesarCITREA/crocus-bert-medical-department-classification')
|
8 |
+
|
9 |
+
# Define the department names
|
10 |
+
departments = [
|
11 |
+
"Kadın Hastalıkları ve Doğum",
|
12 |
+
"Ortopedi ve Travmatoloji" ,
|
13 |
+
"Dermatoloji",
|
14 |
+
"Göğüs Hastalıkları ",
|
15 |
+
"Nöroloji",
|
16 |
+
"Onkoloji" ,
|
17 |
+
"Dahiliye (İç Hastalıkları)" ,
|
18 |
+
"Kardiyoloji",
|
19 |
+
"Psikiyatri" ,
|
20 |
+
"Pediatri" ,
|
21 |
+
"Nefroloji" ,
|
22 |
+
"Fiziksel Tıp ve Rehabilitasyon" ,
|
23 |
+
"Enfeksiyon Hastalıkları ve Klinik Mikrobiyoloji" ,
|
24 |
+
"Üroloji" ,
|
25 |
+
"Kulak Burun Boğaz (KBB)",
|
26 |
+
"Göz Hastalıkları"
|
27 |
+
]
|
28 |
+
|
29 |
+
# Function to predict the department
|
30 |
+
def predict_department(description):
|
31 |
+
# Tokenize input
|
32 |
+
inputs = tokenizer(description, return_tensors="pt", truncation=True, padding=True)
|
33 |
+
|
34 |
+
# Perform inference
|
35 |
+
with torch.no_grad():
|
36 |
+
outputs = model(**inputs)
|
37 |
+
logits = outputs.logits
|
38 |
+
|
39 |
+
# Get the department with the highest score
|
40 |
+
predicted_class = torch.argmax(logits, dim=1).item()
|
41 |
+
|
42 |
+
# Return the department name
|
43 |
+
return departments[predicted_class]
|
44 |
+
|
45 |
+
# Streamlit app interface
|
46 |
+
st.title("Medical Department Classifier")
|
47 |
+
|
48 |
+
# Input text box for the user to describe the symptoms
|
49 |
+
description = st.text_area("Lütfen yaşadığınız tıbbi şikayetleri giriniz:")
|
50 |
+
|
51 |
+
# Button to classify the input
|
52 |
+
if st.button("Classify"):
|
53 |
+
if description:
|
54 |
+
department = predict_department(description)
|
55 |
+
st.write(f"Gitmeniz gereken tıbbi departman: **{department}**")
|
56 |
+
else:
|
57 |
+
st.write("Lütfen yaşadığınız durumu açıklanıyınız.")
|
58 |
+
|
59 |
|
|
|
|