Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,40 +1,179 @@
-import altair as alt
-import numpy as np
-import pandas as pd
 import streamlit as st
+import pandas as pd
+import plotly.express as px
+import json
+import numpy as np
+from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer
 
-"""
-# Welcome to Streamlit!
+# Set Streamlit page configuration
+st.set_page_config(page_title="Construction Materials Dashboard", layout="wide")
 
-Edit `/streamlit_app.py` to customize this app to your heart's desire :heart:.
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
+# Custom CSS for styling
+st.markdown("""
+<style>
+    .navbar {
+        background-color: #1f77b4;
+        padding: 1rem;
+        border-radius: 8px;
+        margin-bottom: 1rem;
+    }
+    .navbar-title {
+        color: white;
+        font-size: 24px;
+        font-weight: bold;
+    }
+    .navbar-links a {
+        color: white;
+        margin-right: 1rem;
+        text-decoration: none;
+        font-size: 16px;
+    }
+    .navbar-links a:hover {
+        text-decoration: underline;
+    }
+    .filter-container {
+        background-color: #f5f5f5;
+        padding: 1rem;
+        border-radius: 8px;
+        margin-bottom: 1rem;
+    }
+</style>
+""", unsafe_allow_html=True)
 
-In the meantime, below is an example of what you can do with just a few lines of code:
-"""
+# Navbar
+st.markdown("""
+<div class="navbar">
+    <div class="navbar-title">Construction Materials Dashboard</div>
+    <div class="navbar-links">
+        <a href="#overview">Overview</a>
+        <a href="#filters">Filters</a>
+        <a href="#insights">LLM Insights</a>
+    </div>
+</div>
+""", unsafe_allow_html=True)
+
+# Load dataset
+@st.cache_data
+def load_data():
+    return pd.read_csv("construction_materials_dataset_3000.csv")
+
+df = load_data()
+
+# Filter section
+st.subheader("Filters", anchor="filters")
+with st.container():
+    st.markdown('<div class="filter-container">', unsafe_allow_html=True)
+    col1, col2, col3 = st.columns(3)
+
+    with col1:
+        product_categories = sorted(df['product_category'].dropna().unique())
+        selected_categories = st.multiselect("Product Category", product_categories, default=product_categories)
+
+    with col2:
+        grades = sorted(df['grade'].dropna().unique())
+        selected_grades = st.multiselect("Grade", grades, default=grades)
+
+    with col3:
+        ratings = sorted(df['ratings'].dropna().astype(str).unique())
+        selected_ratings = st.multiselect("Ratings", ratings, default=ratings)
+
+    st.markdown('</div>', unsafe_allow_html=True)
+
+# Apply filters
+filtered_df = df[
+    (df['product_category'].isin(selected_categories)) &
+    (df['grade'].isin(selected_grades)) &
+    (df['ratings'].astype(str).isin(selected_ratings))
+]
 
-num_points = st.slider("Number of points in spiral", 1, 10000, 1100)
-num_turns = st.slider("Number of turns in spiral", 1, 300, 31)
-
-indices = np.linspace(0, 1, num_points)
-theta = 2 * np.pi * num_turns * indices
-radius = indices
-
-x = radius * np.cos(theta)
-y = radius * np.sin(theta)
-
-df = pd.DataFrame({
-    "x": x,
-    "y": y,
-    "idx": indices,
-    "rand": np.random.randn(num_points),
-})
-
-st.altair_chart(alt.Chart(df, height=700, width=700)
-    .mark_point(filled=True)
-    .encode(
-        x=alt.X("x", axis=None),
-        y=alt.Y("y", axis=None),
-        color=alt.Color("idx", legend=None, scale=alt.Scale()),
-        size=alt.Size("rand", legend=None, scale=alt.Scale(range=[1, 150])),
-    ))
+# Aggregate by supplier
+seller_data = filtered_df.groupby("supplier_name").agg({
+    "bidding_amount": "sum"
+}).reset_index()
+
+# Top 5 sellers
+top_sellers = seller_data.sort_values("bidding_amount", ascending=False).head(5)
+
+# Overview section
+st.subheader("Overview", anchor="overview")
+st.write("**Top 5 Suppliers by Total Bidding Amount**")
+fig_bar = px.bar(
+    top_sellers,
+    x="supplier_name",
+    y="bidding_amount",
+    labels={"supplier_name": "Supplier", "bidding_amount": "Bidding Amount"},
+    title="Top 5 Suppliers",
+    color="supplier_name"
+)
+fig_bar.update_layout(showlegend=False)
+st.plotly_chart(fig_bar, use_container_width=True)
+
+st.write("**Bidding Distribution (Top 5)**")
+fig_pie = px.pie(
+    top_sellers,
+    names="supplier_name",
+    values="bidding_amount",
+    title="Bidding Amount by Supplier"
+)
+st.plotly_chart(fig_pie, use_container_width=True)
+
+total_bidding = seller_data["bidding_amount"].sum()
+st.write(f"**Total Bidding Amount (All Suppliers):** ${total_bidding:,.2f}")
+
+# Convert complex types to simple Python types
+def convert_to_serializable(obj):
+    if isinstance(obj, np.integer):
+        return int(obj)
+    elif isinstance(obj, np.floating):
+        return float(obj)
+    elif isinstance(obj, np.ndarray):
+        return obj.tolist()
+    elif isinstance(obj, (pd.Series, pd.DataFrame)):
+        return obj.to_dict()
+    elif isinstance(obj, dict):
+        return {k: convert_to_serializable(v) for k, v in obj.items()}
+    elif isinstance(obj, list):
+        return [convert_to_serializable(i) for i in obj]
+    return obj
+
+# LLM Section
+st.subheader("Ask Mistral About the Data", anchor="insights")
+user_query = st.text_input("Enter your question:", "Summarize why these are the top 5 suppliers.")
+
+# Load Mistral model from Hugging Face
+@st.cache_resource
+def load_mistral_pipeline():
+    model_id = "mistralai/Mistral-7B-Instruct-v0.1"
+    tokenizer = AutoTokenizer.from_pretrained(model_id)
+    model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype="auto")
+    pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, max_new_tokens=512)
+    return pipe
+
+if user_query:
+    with st.spinner("Generating response..."):
+        pipe = load_mistral_pipeline()
+
+        # Prepare prompt
+        top_sellers_json = json.dumps(convert_to_serializable(top_sellers), indent=2)
+        filters_applied = {
+            "product_category": selected_categories,
+            "grade": selected_grades,
+            "ratings": selected_ratings
+        }
+        prompt = f"""You are a helpful assistant. Based on the dataset below and filters, answer the following user question.
+
+Top 5 sellers:
+{top_sellers_json}
+
+Filters applied:
+{json.dumps(filters_applied, indent=2)}
+
+Question:
+{user_query}
+"""
+        response = pipe(prompt)[0]['generated_text']
+        # Display only the assistant's answer (trim prompt if echoed)
+        st.markdown("**Mistral LLM Response:**")
+        st.write(response.split("Question:")[-1].strip())
+else:
+    st.info("Enter a question to ask Mistral about the bidding data.")