File size: 1,052 Bytes
86c0a21
fe0f1f2
 
86c0a21
fe0f1f2
 
 
 
86c0a21
fe0f1f2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
import streamlit as st
from transformers import AutoModelForSequenceClassification, AutoTokenizer
import torch

# Load the model and tokenizer
model_name = "acorreal/phi3-project-management"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

# Streamlit app
st.title('Project Management Educational Tutor')
st.write('This app uses the "acorreal/phi3-project-management" model')

user_input = st.text_area("Enter your project management question or topic here:")

if st.button('Get Response'):
    if user_input:
        inputs = tokenizer(user_input, return_tensors="pt")
        with torch.no_grad():
            outputs = model(**inputs)
            logits = outputs.logits
            predicted_class_id = logits.argmax().item()

        st.write(f"Predicted class ID: {predicted_class_id}")
        # You can add more logic here to provide detailed responses based on the predicted_class_id
    else:
        st.write("Please enter a question or topic to get a response.")