Change Flatten layer, loss function and add Input

src/constants.py

@@ -3,7 +3,7 @@ TRAIN_DATASET = "data/train_data.tfrecords"
 TEST_DATASET = "data/test_data.tfrecords"
 EVAL_DATASET = "data/eval_data.tfrecords"
 EPOCHS = 1000
-BATCH_SIZE = 256
+BATCH_SIZE = 1
 LEARNING_RATE = 0.004
 L2 = 0.001
 LOG_DIR = "logs"

src/model.py

@@ -1,8 +1,9 @@
 from random import seed
 
-from tensorflow.keras import Model, Sequential, layers
+from tensorflow.keras import Model, Sequential
+from tensorflow.keras.layers import *
 from tensorflow.keras.callbacks import TensorBoard
-from tensorflow.keras.losses import sparse_categorical_crossentropy
+from tensorflow.keras.losses import categorical_crossentropy
 from tensorflow.keras.optimizers import Adam
 from tensorflow.keras.regularizers import l2
 from tensorflow.random import set_seed
@@ -15,47 +16,28 @@ def build_model() -> Model:
     """
     Build the CNN model
     """
-    model = Sequential()
-    model.add(
-        layers.Conv1D(
-            filters=16,
-            kernel_size=5,
-            activation="relu",
-            kernel_regularizer=l2(L2),
-        )
+    model = Sequential(
+        [
+            Input(shape=(None, len(BASES))),
+            Conv1D(
+                filters=16, kernel_size=5, activation="relu", kernel_regularizer=l2(L2)
+            ),
+            MaxPool1D(pool_size=3, strides=1),
+            Conv1D(
+                filters=16, kernel_size=3, activation="relu", kernel_regularizer=l2(L2)
+            ),
+            MaxPool1D(pool_size=3, strides=1),
+            GlobalAveragePooling1D(),
+            Dense(units=16, activation="relu", kernel_regularizer=l2(L2)),
+            Dropout(rate=0.3),
+            Dense(units=16, activation="relu", kernel_regularizer=l2(L2)),
+            Dropout(rate=0.3),
+            Dense(units=len(BASES), activation="softmax"),
+        ]
     )
-    model.add(layers.MaxPool1D(pool_size=3, strides=1))
-    model.add(
-        layers.Conv1D(
-            filters=16,
-            kernel_size=3,
-            activation="relu",
-            kernel_regularizer=l2(L2),
-        )
-    )
-    model.add(layers.MaxPool1D(pool_size=3, strides=1))
-    model.add(layers.Flatten())
-    model.add(
-        layers.Dense(
-            units=16,
-            activation="relu",
-            kernel_regularizer=l2(L2),
-        )
-    )
-    model.add(layers.Dropout(rate=0.3))
-    model.add(
-        layers.Dense(
-            units=16,
-            activation="relu",
-            kernel_regularizer=l2(L2),
-        )
-    )
-    model.add(layers.Dropout(rate=0.3))
-    # FIXME Change output size
-    model.add(layers.Dense(units=len(BASES), activation="softmax"))
     model.compile(
         optimizer=Adam(LEARNING_RATE),
-        loss=sparse_categorical_crossentropy,
+        loss=categorical_crossentropy,
         metrics=["accuracy"],
     )
     return model

src/preprocessing.py

@@ -50,10 +50,7 @@ def read_fastq(data_file, label_file) -> List[bytes]:
     examples = []
     with open(data_file) as data, open(label_file) as labels:
         for element, label in zip(parse(data, "fastq"), parse(labels, "fastq")):
-            example = generate_example(
-                sequence=str(element.seq),
-                label=str(label.seq),
-            )
+            example = generate_example(sequence=str(element.seq), label=str(label.seq))
             examples.append(example)
     return examples