ai-recommendation-engine/dashboard.py at main · gyanpattnaik/ai-recommendation-engine · GitHub

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
import streamlit as st
import pandas as pd
import os

st.set_page_config(page_title="AI Recommendation Engine Dashboard", layout="wide")

BASE_DIR = os.path.dirname(__file__)
REPORTS_DIR = os.path.join(BASE_DIR, 'reports')

@st.cache_data
def load_data():
    results_path = os.path.join(REPORTS_DIR, 'evaluation_results.csv')
    recs_path = os.path.join(REPORTS_DIR, 'ai_recommendations.csv')

    if not os.path.exists(results_path) or not os.path.exists(recs_path):
        return None, None

    results_df = pd.read_csv(results_path)
    recs_df = pd.read_csv(recs_path)
    return results_df, recs_df

st.title("🛍️ AI Recommendation Engine Dashboard")
st.markdown("Interactive comparison of our Semantic AI model vs a simple Popularity Baseline.")

results_df, recs_df = load_data()

if results_df is None:
    st.error("No reports found! Please run the offline evaluation first.")
else:
    # 1. Headline Metrics
    st.header("1. Core Performance Metrics")

    summary = results_df.groupby('model_type')['hit'].mean() * 100
    precision = summary / 5.0

    base_hit = summary.get('baseline', 0)
    ai_hit = summary.get('ai_model', 0)

    col1, col2, col3 = st.columns(3)

    col1.metric("Baseline Hit Rate@5", f"{base_hit:.2f}%")
    col2.metric("AI Model Hit Rate@5", f"{ai_hit:.2f}%", f"+{(ai_hit-base_hit):.2f}% vs Baseline", delta_color="normal")
    col3.metric("AI Model Precision@5", f"{precision.get('ai_model', 0):.2f}%")

    st.divider()

    # 2. Charts
    colA, colB = st.columns(2)

    with colA:
        st.subheader("Category Distribution")
        cat_counts = recs_df['category'].value_counts()
        st.bar_chart(cat_counts, color="#673AB7")

    with colB:
        st.subheader("Hit Rate Trend Line")
        window = st.slider("Smoothing Window Size (Moving Average)", min_value=10, max_value=200, value=50, step=10)

        baseline = results_df[results_df['model_type'] == 'baseline']['hit'].rolling(window=window).mean() * 100
        ai_model = results_df[results_df['model_type'] == 'ai_model']['hit'].rolling(window=window).mean() * 100

        trend_df = pd.DataFrame({
            "AI Model": ai_model.values,
            "Baseline": baseline.values
        })
        st.line_chart(trend_df.dropna(), color=["#FF6F00", "#9E9E9E"])

    st.divider()

    # 3. Explore Raw Data
    st.header("Explore Raw Recommendations")
    user_id = st.selectbox("Select User ID to inspect:", results_df['user_id'].unique())

    user_res = results_df[results_df['user_id'] == user_id]
    st.write("Evaluation Result:")
    st.dataframe(user_res, use_container_width=True, hide_index=True)

    st.write(f"Top 5 Recommendations for {user_id}:")
    user_recs = recs_df[recs_df['user_id'] == user_id]
    st.dataframe(user_recs, use_container_width=True, hide_index=True)