Mnist module


mnist.show_example_image(data: numpy.array, label: numpy.array, true_labels: dict) → None[source]

Prints first image of given data

Parameters:
  • data (np.array) – images
  • label (np.array) – labels of images
  • true_labels (dict) – dictionary with names of labels
Returns:

None

Return type:

None

mnist.get_train_and_test() → tuple[source]

Function for load dataset from pickle file and return test and train set

Returns:tuple of np.arrays
Return type:tuple
mnist.get_dict_labels() → dict[source]
Returns:dictionary with name of labels in mnist dataset
Return type:dict
mnist.main()[source]

Todo

  • TODO: Make stratified train/test split
  • TODO: Stochastic gradien descent and mini batch
  • TODO: Adam solver
  • TODO: Learning rate change during training

Return Home