Parallelize dataset transformations

src/preprocessing.py

@@ -105,8 +105,8 @@ def read_dataset(filepath) -> TFRecordDataset:
     Read TFRecords files and generate a dataset
     """
     data_input = TFRecordDataset(filenames=filepath)
-    shuffled_dataset = dataset.shuffle(buffer_size=10000, seed=42)
     dataset = data_input.map(map_func=process_input, num_parallel_calls=AUTOTUNE)
+    shuffled_dataset = dataset.shuffle(buffer_size=10000, seed=42)
     batched_dataset = shuffled_dataset.batch(batch_size=BATCH_SIZE).repeat(count=EPOCHS)
     return batched_dataset