Create a basic CNN model

src/model.py

@@ -1,11 +1,46 @@
-from tensorflow_io import genome
+from tensorflow.keras import Model, Sequential, layers
-from tensorflow import Tensor
+from tensorflow.keras.regularizers import l2
-from typing import Tuple
+
+from preprocessing import BASES
 
 
-def parse_data(filepath) -> Tuple[Tensor, Tensor]:
+def build_model(hyper_parameters) -> Model:
     """
-    Reads a FASTQ file and returns the sequences and the qualities
+    Builds the CNN model
     """
-    HVR = genome.read_fastq(filename=filepath)
+    return Sequential(
-    return HVR.sequences, HVR.raw_quality
+        [
+            # Two convolutions + maxpooling blocks
+            layers.Conv1D(
+                filters=16,
+                kernel_size=5,
+                activation="relu",
+                kernel_regularizer=l2(hyper_parameters.l2),
+            ),
+            layers.MaxPool1D(pool_size=3, strides=1),
+            layers.Conv1D(
+                filters=16,
+                kernel_size=3,
+                activation="relu",
+                kernel_regularizer=l2(hyper_parameters.l2),
+            ),
+            layers.MaxPool1D(pool_size=3, strides=1),
+            # Flatten the input volume
+            layers.Flatten(),
+            # Two fully connected layers, each followed by a dropout layer
+            layers.Dense(
+                units=16,
+                activation="relu",
+                kernel_regularizer=l2(hyper_parameters.l2),
+            ),
+            layers.Dropout(rate=0.3),
+            layers.Dense(
+                units=16,
+                activation="relu",
+                kernel_regularizer=l2(hyper_parameters.l2),
+            ),
+            layers.Dropout(rate=0.3),
+            # Output layer with softmax activation
+            layers.Dense(units=len(BASES), activation="softmax"),
+        ]
+    )