diff --git a/pipelines/moving_mnist_pipeline/__init__.py b/pipelines/moving_mnist_pipeline/__init__.py
deleted file mode 100644
index e69de29..0000000
diff --git a/pipelines/moving_mnist_pipeline/self_attention_memory_convlstm.py b/pipelines/moving_mnist_pipeline/self_attention_memory_convlstm.py
deleted file mode 100644
index 9155cf2..0000000
--- a/pipelines/moving_mnist_pipeline/self_attention_memory_convlstm.py
+++ /dev/null
@@ -1,99 +0,0 @@
-import os
-
-from torch import nn
-from torch.optim import Adam
-
-from pipelines.evaluator import Evaluator
-from pipelines.moving_mnist_pipeline.data_loader import MovingMNISTDataLoaders
-from pipelines.trainer import Trainer
-from pipelines.utils.early_stopping import EarlyStopping
-from self_attention_memory_convlstm.seq2seq import SAMSeq2Seq
-
-
-def main():
-    ###
-    # Parameters
-    ###
-    train_epochs = 1
-    train_batch_size = 1
-
-    attention_hidden_dims = 1
-    num_channels = 1
-    kernel_size = 3
-    num_kernels = 1
-    padding = "same"
-    activation = "relu"
-    frame_size = (64, 64)
-    num_layers = 1
-    input_seq_length = 10
-    weights_initializer = "he"
-    return_sequences = True
-    out_channels = 1
-
-    ###
-    # DatLoaders
-    ###
-    print("Loading dataset ...")
-    data_loaders = MovingMNISTDataLoaders(
-        train_batch_size, input_frames=input_seq_length
-    )
-
-    ###
-    # Setup training
-    ###
-    loss_criterion = nn.MSELoss()
-    acc_criterion = nn.L1Loss()
-    model = SAMSeq2Seq(
-        attention_hidden_dims=attention_hidden_dims,
-        num_channels=num_channels,
-        kernel_size=kernel_size,
-        num_kernels=num_kernels,
-        padding=padding,
-        activation=activation,
-        frame_size=frame_size,
-        num_layers=num_layers,
-        input_seq_length=input_seq_length,
-        out_channels=out_channels,
-        weights_initializer=weights_initializer,
-        return_sequences=return_sequences,
-    )
-    optimizer = Adam(model.parameters(), lr=0.0005)
-    early_stopping = EarlyStopping(
-        patience=30,
-        verbose=True,
-        delta=0.0001,
-    )
-
-    ###
-    # Training
-    ###
-    print("Training Self Attention (Memory) ConvLSTM...")
-    os.makedirs("./tmp", exist_ok=True)
-    trainer = Trainer(
-        model=model,
-        train_epochs=train_epochs,
-        train_dataloader=data_loaders.train_dataloader,
-        validation_dataloader=data_loaders.validation_dataloader,
-        loss_criterion=loss_criterion,
-        accuracy_criterion=acc_criterion,
-        optimizer=optimizer,
-        early_stopping=early_stopping,
-        artifact_dir="./tmp",
-        metrics_filename="metrics.csv",
-    )
-    trainer.run()
-
-    ###
-    # Evaluation
-    ###
-    print("Evaluating ...")
-    evaluator = Evaluator(
-        model=model,
-        test_dataloader=data_loaders.test_dataloader,
-        artifact_dir="./tmp/evaluate",
-    )
-    evaluator.run()
-
-
-if __name__ == "__main__":
-    main()