File size: 2,591 Bytes
b43c350
 
dff4d95
 
b43c350
 
 
dff4d95
b43c350
 
 
 
 
 
9000775
b43c350
 
 
 
 
 
 
dff4d95
b43c350
 
 
 
 
 
 
 
 
dff4d95
b43c350
 
 
 
 
dff4d95
b43c350
 
dff4d95
b43c350
 
 
 
dff4d95
b43c350
 
 
 
dff4d95
b43c350
 
 
dff4d95
b43c350
 
dff4d95
b43c350
dff4d95
 
b43c350
dff4d95
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
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
from transformers import AutoModelForSequenceClassification, AutoTokenizer
import torch
import gradio as gr

# Load Personality_LM model and tokenizer
model = AutoModelForSequenceClassification.from_pretrained("KevSun/Personality_LM", ignore_mismatched_sizes=True)
tokenizer = AutoTokenizer.from_pretrained("KevSun/Personality_LM")

def analyze_personality(text):
    """Analyze personality traits from input text."""
    encoded_input = tokenizer(text, return_tensors='pt', padding=True, truncation=True, max_length=512)
    model.eval()
    with torch.no_grad():
        outputs = model(**encoded_input)

    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
    predicted_scores = predictions[0].tolist()
    
    trait_names = ["agreeableness", "openness", "conscientiousness", "extraversion", "neuroticism"]
    personality_traits = {trait: score for trait, score in zip(trait_names, predicted_scores)}

    return personality_traits

def adjust_response(response, traits):
    """Adjust chatbot response based on personality traits."""
    if traits["agreeableness"] > 0.5:
        response = f"{response} 😊 I'm so glad we get along well!"
    if traits["neuroticism"] > 0.5:
        response += " But I'm feeling a bit worried about what might happen..."
    if traits["extraversion"] > 0.5:
        response += " Let's keep chatting! I love interacting with you."
    return response

def respond(user_message, history, personality_text):
    """Generate chatbot response based on user input and personality."""
    traits = analyze_personality(personality_text)
    base_response = f"Hi! You said: {user_message}"
    final_response = adjust_response(base_response, traits)

    history.append((user_message, final_response))
    return history, history

def personality_demo():
    """Create the Gradio interface for the chatbot with personality training."""
    with gr.Blocks() as demo:
        gr.Markdown("### Personality-Based Chatbot")

        personality_textbox = gr.Textbox(
            label="Define Personality Text (Use direct input if no file)", 
            placeholder="Type personality description or paste a sample text here."
        )

        chatbot = gr.Chatbot()
        msg = gr.Textbox(label="User Input", placeholder="Say something to the chatbot...")
        clear = gr.Button("Clear Chat")

        msg.submit(respond, [msg, chatbot, personality_textbox], [chatbot, chatbot])
        clear.click(lambda: ([], []), None, [chatbot, chatbot])

    return demo

if __name__ == "__main__":
    demo = personality_demo()
    demo.launch()