File size: 6,499 Bytes
28635a8
 
 
 
 
 
 
 
 
 
 
 
7918fa4
28635a8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f0d7b32
28635a8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
440bd6b
28635a8
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
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
import gradio as gr
from sentence_transformers import SentenceTransformer, util
import openai
import os
import os
os.environ["TOKENIZERS_PARALLELISM"] = "false"


# Initialize paths and model identifiers for easy configuration and maintenance
filename = "output_country_details.txt"  # Path to the file storing country-specific details
retrieval_model_name = 'output/sentence-transformer-finetuned/'

openai.api_key = os.environ["OPENAI_API_KEY"]


# Attempt to load the necessary models and provide feedback on success or failure
try:
    retrieval_model = SentenceTransformer(retrieval_model_name)
    print("Models loaded successfully.")
except Exception as e:
    print(f"Failed to load models: {e}")

def load_and_preprocess_text(filename):
    """
    Load and preprocess text from a file, removing empty lines and stripping whitespace.
    """
    try:
        with open(filename, 'r', encoding='utf-8') as file:
            segments = [line.strip() for line in file if line.strip()]
        print("Text loaded and preprocessed successfully.")
        return segments
    except Exception as e:
        print(f"Failed to load or preprocess text: {e}")
        return []

segments = load_and_preprocess_text(filename)

def find_relevant_segment(user_query, segments):
    """
    Find the most relevant text segment for a user's query using cosine similarity among sentence embeddings.
    This version tries to match country names in the query with those in the segments.
    """
    try:
        # Lowercase the query for better matching
        lower_query = user_query.lower()
        # Filter segments to include only those containing country names mentioned in the query
        country_segments = [seg for seg in segments if any(country.lower() in seg.lower() for country in ['Guatemala', 'Mexico', 'U.S.', 'United States'])]
        
        # If no specific country segments found, default to general matching
        if not country_segments:
            country_segments = segments
        
        query_embedding = retrieval_model.encode(lower_query)
        segment_embeddings = retrieval_model.encode(country_segments)
        similarities = util.pytorch_cos_sim(query_embedding, segment_embeddings)[0]
        best_idx = similarities.argmax()
        return country_segments[best_idx]
    except Exception as e:
        print(f"Error in finding relevant segment: {e}")
        return ""


def generate_response(user_query, relevant_segment):
    """
    Generate a response emphasizing the bot's capability in providing country-specific visa information.
    """
    try:
        system_message = "You are a visa chatbot specialized in providing country-specific visa requirement information."
        user_message = f"Here's the information on visa requirements for your query: {relevant_segment}"
        messages = [
            {"role": "system", "content": system_message},
            {"role": "user", "content": user_message}
        ]
        response = openai.ChatCompletion.create(
            model="gpt-3.5-turbo",  # Verify model name
            messages=messages,
            max_tokens=150,
            temperature=0.2,
            top_p=1,
            frequency_penalty=0,
            presence_penalty=0
        )
        return response['choices'][0]['message']['content'].strip()
    except Exception as e:
        print(f"Error in generating response: {e}")
        return f"Error in generating response: {e}"




# Define and configure the Gradio application interface to interact with users.
# Define and configure the Gradio application interface to interact with users.
def query_model(question):
    """
    Process a question, find relevant information, and generate a response, specifically for U.S. visa questions.
    """
    if question == "":
        return "Welcome to VisaBot! Ask me anything about U.S. visa processes."
    relevant_segment = find_relevant_segment(question, segments)
    if not relevant_segment:
        return "Could not find U.S.-specific information. Please refine your question."
    response = generate_response(question, relevant_segment)
    return response




# Define the welcome message and specific topics and countries the chatbot can provide information about.
welcome_message = """
# Welcome to VISABOT!

## Your AI-driven visa assistant for all travel-related queries.
"""

topics = """
### Feel Free to ask me anything from the topics below!
- Visa issuance
- Documents needed
- Application process
- Processing time
- Recommended Vaccines
- Health Risks
- Healthcare Facilities
- Currency Information
- Embassy Information 
- Allowed stay
"""

countries = """
### Our chatbot can currently answer questions for these countries!
- πŸ‡¨πŸ‡³ China
- πŸ‡«πŸ‡· France
- πŸ‡¬πŸ‡Ή Guatemala
- πŸ‡±πŸ‡§ Lebanon
- πŸ‡²πŸ‡½ Mexico
- πŸ‡΅πŸ‡­ Philippines
- πŸ‡·πŸ‡Έ Serbia
- πŸ‡ΈπŸ‡± Sierra Leone
- πŸ‡ΏπŸ‡¦ South Africa
- πŸ‡»πŸ‡³ Vietnam
"""

# Define and configure the Gradio application interface to interact with users.
def query_model(question):
    """
    Process a question, find relevant information, and generate a response.
    
    Args:
        question (str): User's input question.

    Returns:
        str: Generated response or a default welcome message if no question is provided.
    """
    if question == "":
        return welcome_message
    relevant_segment = find_relevant_segment(question, segments)
    response = generate_response(question, relevant_segment)
    return response




# Setup the Gradio Blocks interface with custom layout components
with gr.Blocks() as demo:
    gr.Markdown(welcome_message)  # Display the formatted welcome message
    with gr.Row():
        with gr.Column():
            gr.Markdown(topics)  # Show the topics on the left side
        with gr.Column():
            gr.Markdown(countries)  # Display the list of countries on the right side
    with gr.Row():
        img = gr.Image(os.path.join(os.getcwd(), "final.png"), width=500)  # Include an image for visual appeal
    with gr.Row():
        with gr.Column():
            question = gr.Textbox(label="Your question", placeholder="What do you want to ask about?")
            answer = gr.Textbox(label="VisaBot Response", placeholder="VisaBot will respond here...", interactive=False, lines=10)
            submit_button = gr.Button("Submit")
            submit_button.click(fn=query_model, inputs=question, outputs=answer)

# Launch the Gradio app to allow user interaction
demo.launch(share=True)