Stuck here while using Keras (2.3.1)

Question

This is my DataSet

import keras

print(keras.__version__)

mnist = keras.datasets.mnist

(x_train,y_train),(x_test,y_test) = mnist.load_data()

I have performed normalizing

x_train = keras.utils.normalize(x_train,axis=1)

x_test = keras.utils.normalize(x_test,axis=1)

And built a model

model = keras.models.Sequential()

model.add(keras.layers.Flatten(x_train))

model.add(keras.layers.Dense(128,activation= keras.nn.relu))

model.add(keras.layers.Dense(128,activation= keras.nn.relu))

model.add(keras.layers.Dense(10,activation= keras.nn.softmax))

model.compile(optimizer='adam',

              loss='sparse_categorical_crossentropy',

              metrics = ['accuracy']

)

model.fit(x_train,y_train,epochs=3)

I am getting the below error:

Using TensorFlow backend.

2.3.1

Traceback (most recent call last):

  File "/Users/aditya/Desktop/Desktop/dataScience/Practice/OpenCV/FaceDetect/Hackathon/classMnist.py", line 28, in <module>

    model.add(keras.layers.Flatten(x_train))

  File "/usr/local/lib/python3.7/site-packages/keras/layers/core.py", line 495, in __init__

    self.data_format = K.normalize_data_format(data_format)

  File "/usr/local/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py", line 311, in normalize_data_format

    data_format = value.lower()

AttributeError: 'numpy.ndarray' object has no attribute 'lower'

The problem is that the Keras can't flatten my  x_train dataset. Do you have nay idea why is this hapenning?

Answer 1

model.add(keras.layers.Flatten(x_train))

The Keras create the network that can't load data.

model.fit(x_train,y_train,epochs=3)

The data gets loaded.

So you should first edit the code:

model.add(keras.layers.Flatten())

And your codes have other error to:

# wrong

model.add(keras.layers.Dense(128,activation= keras.nn.relu))

# right

model.add(keras.layers.Dense(128,activation= keras.backend.relu))

If you want to know more about the Data Science then do check out the following Data Science which will help you in understanding Data Science from scratch