add CNN model — model selector, Conv1DNetwork class, dual model loading

src/model_def.py

@@ -1,7 +1,7 @@
-"""FeedForwardNetwork definition required for deserializing model.keras.
+"""Model classes required for deserializing .keras files.
 
-This must be imported before tf.keras.models.load_model() is called so
-that keras can resolve the registered custom class.
+Must be imported before tf.keras.models.load_model() so that keras
+can resolve the registered custom classes.
 """
 
 import keras
@@ -63,3 +63,90 @@ class FeedForwardNetwork(tf.keras.Model):
             "dropout_rate": self._dropout_rate,
         })
         return config
+
+
+@keras.saving.register_keras_serializable()
+class Conv1DNetwork(tf.keras.Model):
+    """1D-CNN for classification on pre-computed feature vectors.
+
+    Architecture: Reshape(561, 1)
+                  → Conv1D(64,  k=5) → BN → ReLU → MaxPool(2) → Dropout(0.3)
+                  → Conv1D(128, k=5) → BN → ReLU → MaxPool(2) → Dropout(0.3)
+                  → Conv1D(256, k=3) → BN → ReLU → GlobalAvgPool1D
+                  → Dense(128) → BN → ReLU → Dropout(0.5)
+                  → Dense(6, softmax)
+    """
+
+    def __init__(
+        self,
+        num_features,
+        num_classes,
+        dropout_rate=0.3,
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+        self._num_features = num_features
+        self._num_classes = num_classes
+        self._dropout_rate = dropout_rate
+
+        self.reshape = tf.keras.layers.Reshape((num_features, 1))
+
+        self.conv1 = tf.keras.layers.Conv1D(64, kernel_size=5, padding="same", use_bias=False)
+        self.bn1   = tf.keras.layers.BatchNormalization()
+        self.relu1 = tf.keras.layers.ReLU()
+        self.pool1 = tf.keras.layers.MaxPooling1D(pool_size=2)
+        self.drop1 = tf.keras.layers.Dropout(dropout_rate)
+
+        self.conv2 = tf.keras.layers.Conv1D(128, kernel_size=5, padding="same", use_bias=False)
+        self.bn2   = tf.keras.layers.BatchNormalization()
+        self.relu2 = tf.keras.layers.ReLU()
+        self.pool2 = tf.keras.layers.MaxPooling1D(pool_size=2)
+        self.drop2 = tf.keras.layers.Dropout(dropout_rate)
+
+        self.conv3 = tf.keras.layers.Conv1D(256, kernel_size=3, padding="same", use_bias=False)
+        self.bn3   = tf.keras.layers.BatchNormalization()
+        self.relu3 = tf.keras.layers.ReLU()
+        self.gap   = tf.keras.layers.GlobalAveragePooling1D()
+
+        self.dense1  = tf.keras.layers.Dense(128, use_bias=False)
+        self.bn_fc   = tf.keras.layers.BatchNormalization()
+        self.relu_fc = tf.keras.layers.ReLU()
+        self.drop_fc = tf.keras.layers.Dropout(0.5)
+
+        self.output_layer = tf.keras.layers.Dense(num_classes, activation="softmax")
+
+    def call(self, inputs, training=False):
+        x = self.reshape(inputs)
+
+        x = self.conv1(x)
+        x = self.bn1(x, training=training)
+        x = self.relu1(x)
+        x = self.pool1(x)
+        x = self.drop1(x, training=training)
+
+        x = self.conv2(x)
+        x = self.bn2(x, training=training)
+        x = self.relu2(x)
+        x = self.pool2(x)
+        x = self.drop2(x, training=training)
+
+        x = self.conv3(x)
+        x = self.bn3(x, training=training)
+        x = self.relu3(x)
+        x = self.gap(x)
+
+        x = self.dense1(x)
+        x = self.bn_fc(x, training=training)
+        x = self.relu_fc(x)
+        x = self.drop_fc(x, training=training)
+
+        return self.output_layer(x)
+
+    def get_config(self):
+        config = super().get_config()
+        config.update({
+            "num_features": self._num_features,
+            "num_classes": self._num_classes,
+            "dropout_rate": self._dropout_rate,
+        })
+        return config

src/streamlit_app.py

@@ -6,7 +6,6 @@ import pandas as pd
 # Paths anchored to the repo root regardless of working directory
 _SRC_DIR = os.path.dirname(os.path.abspath(__file__))   # /app/src
 _REPO_ROOT = os.path.dirname(_SRC_DIR)                  # /app
-_MODEL_PATH = os.path.join(_REPO_ROOT, "model.keras")
 _SAMPLES_PATH = os.path.join(_REPO_ROOT, "data", "samples.csv")
 
 # ── Constants ──────────────────────────────────────────────────────────────
@@ -29,23 +28,31 @@ EXPLANATIONS = {
     "WALKING_UPSTAIRS": "Elevated vertical acceleration effort with upward body displacement — consistent with climbing stairs.",
 }
 
+MODEL_FILES = {
+    "FFN  (512→256→128)": "model.keras",
+    "CNN  (Conv1D×3)":    "har_cnn.keras",
+}
+
 # ── Model loader ────────────────────────────────────────────────────────────
 
 @st.cache_resource
-def load_model():
+def load_model(filename: str):
     try:
         from huggingface_hub import hf_hub_download
         import tensorflow as tf
-        from model_def import FeedForwardNetwork
+        from model_def import FeedForwardNetwork, Conv1DNetwork  # noqa: F401 — registers both classes
 
         model_path = hf_hub_download(
             repo_id="Group3DActRecog/actRecog",
-            filename="model.keras",
+            filename=filename,
             repo_type="space",
         )
         model = tf.keras.models.load_model(
             model_path,
-            custom_objects={"FeedForwardNetwork": FeedForwardNetwork},
+            custom_objects={
+                "FeedForwardNetwork": FeedForwardNetwork,
+                "Conv1DNetwork": Conv1DNetwork,
+            },
         )
         return model, "ready"
     except Exception as e:
@@ -79,15 +86,18 @@ with st.sidebar:
     **Classes:** 6 activities of daily living
     """)
     st.markdown("---")
-    st.markdown("**Model performance on test set**")
-    st.metric("Architecture", "FFN 512→256→128")
-    st.metric("Status", "FFN live · CNN pending")
+    st.markdown("**Select model**")
+    model_choice = st.radio(
+        label="model",
+        options=list(MODEL_FILES.keys()),
+        label_visibility="collapsed",
+    )
     st.markdown("---")
     st.caption("DAT606 Group Assignment · Pan-Atlantic University")
 
-# ── Model status ─────────────────────────────────────────────────────────────
+# ── Load selected model ───────────────────────────────────────────────────────
 
-model, model_status = load_model()
+model, model_status = load_model(MODEL_FILES[model_choice])
 
 if model_status != "ready":
     st.warning(f"Model not loaded — {model_status}")
@@ -130,7 +140,7 @@ with tab1:
             st.metric("Feature count", len(feature_vector))
 
         if st.button("Classify this sample", type="primary"):
-            if model_status == "no_model":
+            if model_status != "ready":
                 st.error("Model not loaded — cannot predict yet.")
             else:
                 arr = feature_vector.reshape(1, -1)