Hey guys I do need you! - For some reason, my model is not satisfying the test cases

TensorFlow General Discussion
,
1 
df = pd.read_csv('Weekly_U.S.Diesel_Retail_Prices.csv',
                     infer_datetime_format=True, index_col='Week of', header=0)

    
N_FEATURES = len(df.columns) 


data = df.values
data = normalize_series(data, data.min(axis=0), data.max(axis=0))


SPLIT_TIME = int(len(data) * 0.8)  
x_train = data[:SPLIT_TIME]
x_valid = data[SPLIT_TIME:]

tf.keras.backend.clear_session()
tf.random.set_seed(42)


BATCH_SIZE = 32  
N_PAST = 10  
N_FUTURE = 10  


SHIFT = 1 


train_set = windowed_dataset(series=x_train, batch_size=BATCH_SIZE,
                              n_past=N_PAST, n_future=N_FUTURE,
                              shift=SHIFT)
valid_set = windowed_dataset(series=x_valid, batch_size=BATCH_SIZE,
                              n_past=N_PAST, n_future=N_FUTURE,
                              shift=SHIFT)


model = tf.keras.models.Sequential([

    tf.keras.layers.InputLayer(input_shape=(N_PAST, N_FEATURES),name='Input'),
    tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(N_FEATURES, input_shape=(BATCH_SIZE,N_PAST, N_FEATURES),activation='relu', return_sequences=True)),
    tf.keras.layers.Dense(N_FEATURES)
])


model.compile(
    
    loss='mae',
    optimizer=tf.keras.optimizers.Adam(),
    metrics=['mae']
)
model.fit(
    
    train_set,
    epochs = 50,

)
3

@Geoffrey,

Welcome to the Tensorflow Forum!

For some reason, my model is not satisfying the test cases

Can you please elaborate on the above statement?

5

I am having the same problem.