Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,40 +1,257 @@
-import altair as alt
-import numpy as np
-import pandas as pd
 import streamlit as st
+import pandas as pd
+import numpy as np
+import altair as alt
+import os
+import time
+import joblib
+
+# =========================
+# PATH SETUP
+# =========================
+BASE_DIR = os.path.dirname(os.path.dirname(__file__))
+
+DATA_PATH = os.path.join(BASE_DIR, "val.csv")
+
+REG_PATH = os.path.join(BASE_DIR, "model_trainer/regression/model.pkl")
+CLF_PATH = os.path.join(BASE_DIR, "model_trainer/classification/model.pkl")
+CLUSTER_PATH = os.path.join(BASE_DIR, "model_trainer/clustering/model.pkl")
+SCALER_PATH = os.path.join(BASE_DIR, "model_trainer/clustering/scaler.pkl")
+
+# =========================
+# LOAD MODELS
+# =========================
+@st.cache_resource
+def load_models():
+    reg = joblib.load(REG_PATH)
+    clf = joblib.load(CLF_PATH)
+    cluster = joblib.load(CLUSTER_PATH)
+    scaler = joblib.load(SCALER_PATH)
+    return reg, clf, cluster, scaler
+
+reg_model, clf_model, cluster_model, cluster_scaler = load_models()
+
+# =========================
+# LOAD DATA
+# =========================
+@st.cache_data
+def load_data():
+    return pd.read_csv(DATA_PATH)
+
+FULL_DATA = load_data()
+
+# =========================
+# STREAM STATE
+# =========================
+if "cursor" not in st.session_state:
+    st.session_state.cursor = 0
+
+if "telemetry" not in st.session_state:
+    st.session_state.telemetry = pd.DataFrame()
+
+# =========================
+# UI CONFIG
+# =========================
+st.set_page_config(page_title="Race Telemetry", layout="wide")
+
+st.markdown("""
+<style>
+.ml-label {
+    font-size: 20px;
+    color: #94a3b8;
+    margin-bottom: 0px;
+}
+.ml-value {
+    font-size: 38px;       
+    font-weight: 500;
+    line-height: 2;
+}
+</style>
+""", unsafe_allow_html=True)
+
+# =========================
+# SIDEBAR
+# =========================
+st.sidebar.title("Pit Wall Controls")
+
+auto_refresh = st.sidebar.toggle("Auto Refresh", value=True)
+refresh_interval = st.sidebar.slider("Refresh Interval (seconds)", 1, 5, 1)
+batch_size = st.sidebar.selectbox("Rows per fetch", [1, 5, 10], index=0)
+
+# =========================
+# FETCH STREAM DATA
+# =========================
+def fetch_rows(batch_size):
+    start = st.session_state.cursor
+    end = start + batch_size
+
+    batch = FULL_DATA.iloc[start:end].copy()
+    st.session_state.cursor = end
+
+    return batch
+
+# =========================
+# REAL INFERENCE
+# =========================
+def run_inference(df_batch):
+    outputs = []
+
+    for _, row in df_batch.iterrows():
+        row_dict = row.to_dict()
+
+        # -------- REGRESSION --------
+        reg_features = ["speed", "current_engine_rpm", "boost", "torque"]
+        X_reg = np.array([row_dict[f] for f in reg_features]).reshape(1, -1)
+        pred_lap = float(reg_model.predict(X_reg)[0])
+
+        # -------- CLASSIFICATION --------
+        clf_features = ["speed", "current_engine_rpm", "gear"]
+        X_clf = np.array([row_dict[f] for f in clf_features]).reshape(1, -1)
+        pred_gear = int(clf_model.predict(X_clf)[0])
+
+        # -------- CLUSTERING --------
+        cluster_features = ["speed", "current_engine_rpm", "boost", "torque", "avg_tire_temp"]
+        X_cluster = np.array([row_dict[f] for f in cluster_features]).reshape(1, -1)
+        X_scaled = cluster_scaler.transform(X_cluster)
+
+        label = cluster_model.predict(X_scaled)[0]
+        behavior = "Aggressive" if label == 1 else "Smooth"
+
+        # attach predictions
+        row_dict["predicted_lap_time"] = pred_lap
+        row_dict["predicted_gear"] = pred_gear
+        row_dict["driving_behavior"] = behavior
+
+        outputs.append(row_dict)
+
+    return pd.DataFrame(outputs)
+
+# =========================
+# FETCH + PROCESS
+# =========================
+new_data = fetch_rows(batch_size)
+
+if not new_data.empty:
+    processed = run_inference(new_data)
+
+    st.session_state.telemetry = pd.concat(
+        [st.session_state.telemetry, processed],
+        ignore_index=True
+    )
+
+df = st.session_state.telemetry
+
+if df.empty:
+    st.stop()
+
+df["t"] = range(len(df))
+latest = df.iloc[-1]
+
+# =========================
+# TITLE
+# =========================
+st.markdown(
+    """
+    <div style="text-align:center; line-height:0;">
+        <h2>🏁 Race Telemetry</h2>
+        <h4>Pit Wall Dashboard</h4>
+    </div>
+    """,
+    unsafe_allow_html=True
+)
+
+# =========================
+# MAIN GRID
+# =========================
+left, right = st.columns([3, 1])
+
+with left:
+
+    with st.container(border=True):
+        c1, c2, c3 = st.columns(3)
+
+        with c1:
+            st.markdown('<div class="ml-label">Predicted Lap</div>', unsafe_allow_html=True)
+            st.markdown(
+                f'<div class="ml-value" style="color:#38bdf8;">{latest["predicted_lap_time"]:.2f} s</div>',
+                unsafe_allow_html=True
+            )
+
+        with c2:
+            st.markdown('<div class="ml-label">Recommended Gear</div>', unsafe_allow_html=True)
+            st.markdown(
+                f'<div class="ml-value" style="color:#22c55e;">{int(latest["predicted_gear"])}</div>',
+                unsafe_allow_html=True
+            )
+
+        with c3:
+            st.markdown('<div class="ml-label">Driving Style</div>', unsafe_allow_html=True)
+            st.markdown(
+                f'<div class="ml-value" style="color:#facc15;">{latest["driving_behavior"]}</div>',
+                unsafe_allow_html=True
+            )
+
+    st.markdown("<br>", unsafe_allow_html=True)
+
+    r2c1, r2c2 = st.columns(2)
+
+    r2c1.metric("Speed (km/h)", f"{latest['speed']:.1f}")
+    r2c1.altair_chart(
+        alt.Chart(df).mark_line().encode(x="t:Q", y="speed:Q"),
+        use_container_width=True
+    )
+
+    r2c2.metric("Engine RPM", int(latest["current_engine_rpm"]))
+    r2c2.altair_chart(
+        alt.Chart(df).mark_area(opacity=0.7).encode(x="t:Q", y="current_engine_rpm:Q"),
+        use_container_width=True
+    )
+
+with right:
+    st.markdown("### Track")
+    st.image("assets/track.png", use_container_width=True)
+
+# =========================
+# LOWER METRICS
+# =========================
+p1, p2, p3, p4 = st.columns(4)
+
+df["power_kw"] = df["power"] / 1000
+
+p1.metric("Power (kW)", f"{df['power_kw'].iloc[-1]:.1f}")
+p1.altair_chart(alt.Chart(df).mark_area().encode(x="t", y="power_kw"), use_container_width=True)
+
+p2.metric("Torque (Nm)", f"{latest['torque']:.1f}")
+p2.altair_chart(alt.Chart(df).mark_line().encode(x="t", y="torque"), use_container_width=True)
+
+p3.metric("Boost (psi)", f"{latest['boost']:.2f}")
+p3.altair_chart(alt.Chart(df).mark_line().encode(x="t", y="boost"), use_container_width=True)
+
+p4.metric("Avg Tire Temp (°C)", f"{latest['avg_tire_temp']:.1f}")
+p4.altair_chart(alt.Chart(df).mark_line().encode(x="t", y="avg_tire_temp"), use_container_width=True)
+
+# =========================
+# ATTITUDE
+# =========================
+attitude_chart = alt.Chart(df).transform_fold(
+    ["yaw", "pitch", "roll"],
+    as_=["Axis", "Value"]
+).mark_line().encode(
+    x="t:Q",
+    y="Value:Q",
+    color="Axis:N"
+)
+
+st.metric(
+    "Yaw / Pitch / Roll (rad)",
+    f"{latest['yaw']:.2f}, {latest['pitch']:.2f}, {latest['roll']:.2f}"
+)
+
+st.altair_chart(attitude_chart, use_container_width=True)
 
-"""
-# Welcome to Streamlit!
-
-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).
-
-In the meantime, below is an example of what you can do with just a few lines of code:
-"""
-
-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])),
-    ))
+# =========================
+# AUTO REFRESH
+# =========================
+if auto_refresh:
+    time.sleep(refresh_interval)
+    st.rerun()