A simple CNN for the MNIST datasets – II – building the CNN with Keras and a first test

I continue with my series on first exploratory steps with CNNs. After all the discussion of CNN basics in the last article,

A simple CNN for the MNIST datasets – I,

we are well prepared to build a very simple CNN with Keras. By simple I mean simple enough to handle the MNIST digit images. The Keras API for creating CNN models, layers and activation functions is very convenient; a simple CNN does not require much code. So, the Jupyter environment is sufficient for our first experiment.

An interesting topic is the use of a GPU. After a somewhat frustrating experience with a MLP on the GPU of a NV 960 GTX in comparison to a i7 6700K CPU I am eager to see whether we get a reasonable GPU acceleration for a CNN. So, we should prepare our code to use the GPU. This requires a bit of preparation.

We should also ask a subtle question: What do we expect from a CNN in comparison to a MLP regarding the MNIST data? A MLP with 2 hidden layers (with 70 and 30 nodes) provided over 99.5% accuracy on the training data and almost 98% accuracy on a test dataset after some tweaking. Even with basic settings for our MLP we arrived at a value over 97.7% after 35 epochs – below 8 secs. Well, a CNN is probably better in feature recognition than a cluster detection algorithm. But we are talking about the last 2 % of remaining accuracy. I admit that I did not know what to expect …

A MLP as an important part of a CNN

At the end of the last article I had discussed a very simple layer structure of convolutional and pooling layers:

  • Layer 0: Input layer (tensor of original image data, 3 layers for color channels or one layer for a gray channel)
  • Layer 1: Conv layer (small 3×3 kernel, stride 1, 32 filters => 32 maps (26×26), overlapping filter areas)
  • Layer 2: Pooling layer (2×2 max pooling => 32 (13×13) maps,
    a map node covers 4×4 non overlapping areas per node on the original image)
  • Layer 3: Conv layer (3×3 kernel, stride 1, 64 filters => 64 maps (11×11),
    a map node covers 8×8 overlapping areas on the original image (total effective stride 2))
  • Layer 4: Pooling layer (2×2 max pooling => 64 maps (5×5),
    a map node covers 10×10 areas per node on the original image (total effective stride 5), some border info lost)
  • Layer 5: Conv layer (3×3 kernel, stride 1, 64 filters => 64 maps (3×3),
    a map node covers 18×18 areas per node (effective stride 5), some border info lost )

This is the CNN structure we are going to use in the near future. (Actually, I followed a suggestion of Francois Chollet; see the literature list in the last article). Let us assume that we somehow have established all these convolution and pooling layers for a CNN. Each layer producse some “feature“-related output, structured in form of a tensors. This led to an open question at the end of the last article:

Where and by what do we get a classification of the resulting data with respect to the 10 digit categories of the MNIST images?

Applying filters and extracting “feature hierarchies” of an image alone does not help without a “learned” judgement about these data. But the answer is very simple:

Use a MLP after the last Conv layer and feed it with data from this Conv layer!

When we think in terms of nodes and artificial neurons, we could say: We just have to connect the “nodes” of the feature maps of layer 5
in our special CNN with the nodes of an input layer of a MLP. As a MLP has a flat input layer we need to prepare 9×64 = 576 receiving “nodes” there. We would use weights with a value of “1.0” along these special connections.

Mathematically, this approach can be expressed in terms of a “flattening” operation on the tensor data produced by the the last Conv data. In Numpy terms: We need to reshape the multidimensional tensor containing the values across the stack of maps at the last Conv2D layer into a long 1D array (= a vector).

From a more general perspective we could say: Feeding the output of the Conv part of our CNN into a MLP for classification is quite similar to what we did when we pre-processed the MNIST data by an unsupervised cluster detection algorithm; also there we used the resulting data as input to an MLP. There is one big difference, however:

The optimization of the network’s weights during training requires a BW propagation of error terms (more precisely: derivatives of the CNN’s loss function) across the MLP AND the convolutional and pooling layers. Error BW propagation should not be stopped at the MLP’s input layer: It has to move from the output layer of the MLP back to the MLP’s input layer and from there to the convolutional and pooling layers. Remember that suitable filter kernels have to be found during (supervised) training.

If you read my PDF on the error back propagation for a MLP
PDF on the math behind Error Back_Propagation
and think a bit about its basic recipes and results you quickly see that the “input layer” of the MLP is no barrier to error back propagation: The “deltas” discussed in the PDF can be back-propagated right down to the MLP’s input layer. Then we just apply the chain rule again. The partial derivatives at the nodes of the input layer with respect to their input values are just “1”, because the activation function there is the identity function. The “weights” between the last Conv layer and the input layer of the MLP are no free parameters – we do not need to care about them. And then everything goes its normal way – we apply chain rule after chain rule for all nodes of the maps to determine the gradients of the CNN’s loss function with respect to the weights there. But you need not think about the details – Keras and TF2 will take proper care about everything …

But, you should always keep the following in mind: Whatever we discuss in terms of layers and nodes – in a CNN these are only fictitious interpretations of a series of mathematical operations on tensor data. Not less, not more ..,. Nodes and layers are just very helpful (!) illustrations of non-cyclic graphs of mathematical operations. KI on the level of my present discussion (MLPs, CNNs) “just” corresponds to algorithms which emerge out of a specific deterministic approach to solve an optimization problem.

Using Tensorflow 2 and Keras

Let us now turn to coding. To be able to use a Nvidia GPU we need a Cuda/Cudnn installation and a working Tensorflow backend for Keras. I have already described the installation of CUDA 10.2 and CUDNN on an Opensuse Linux system in some detail in the article Getting a Keras based MLP to run with Cuda 10.2, Cudnn 7.6 and TensorFlow 2.0 on an Opensuse Leap 15.1 system. You can follow the hints there. In case you run into trouble on your Linux distribution try everything with Cuda 10.1.

Some more hints: TF2 in version 2.2 can be installed by the Pip-mechanism in your virtual Python environment (“pip install –upgrade tensorflow”). TF2 contains already a special Keras version – which is the one we are going to use in our upcoming experiment. So, there is no need to install Keras separately with “pip”. Note also that, in contrast to TF1, it is NOT necessary to install a separate package “tensorflow-gpu”. If all these things are new to you: You find some articles on creating an adequate ML test and development environment based on Python/PyDev/Jupyter somewhere else in this blog.

Imports and settings for CPUs/GPUs

We shall use a Jupyter notebook to perform the basic experiments; but I recommend strongly to consolidate your code in Python files of an Eclipse/PyDev environment afterwards. Before you start your virtual Python environment from a Linux shell you should set the following environment variables:

$>export OPENBLAS_NUM_THREADS=4 # or whatever is reasonable for your CPU (but do not use all CPU cores and/or hyper threads                            
$>export OMP_NUM_THREADS=4                                
$>export TF_XLA_FLAGS=--tf_xla_cpu_global_jit
$>export XLA_FLAGS=--xla_gpu_cuda_data_dir=/usr/local/cuda
$>source bin/activate                                     
(ml_1) $> jupyter notebook

Required Imports

The following commands in a first Jupyter cell perform the required library imports:

import numpy as np
import scipy
import time 
import sys 
import os

import tensorflow as tf
from tensorflow import keras as K
from tensorflow.python.keras import backend as B 
from keras import models
from keras import layers
from keras.utils import to_categorical
from keras.datasets import mnist
from tensorflow.python.client import device_lib

from sklearn.preprocessing import StandardScaler

Do not ignore the statement “from tensorflow.python.keras import backend as B“; we need it later.

The “StandardScaler” of Scikit-Learn will help us to “standardize” the MNIST input data. This is a step which you should know already from MLPs … You can, of course, also experiment with different normalization procedures. But in my opinion using the “StandardScaler” is just convenient. ( I assume that you already have installed scikit-learn in your virtual Python environment).

Settings for CPUs/GPUs

With TF2 the switching between CPU and GPU is a bit of a mess. Not all new parameters and their settings work as expected. As I have explained in the article on the Cuda installation named above, I, therefore, prefer to an old school, but reliable TF1 approach and use the compatibility interface:

#gpu = False 
gpu = True
if gpu: 
    GPU = True;  CPU = False; num_GPU = 1; num_CPU = 1
else: 
    GPU = False; CPU = True;  num_CPU = 1; num_GPU = 0

config = tf.compat.v1.ConfigProto(intra_op_parallelism_threads=6,
                        inter_op_parallelism_threads=1, 
                        allow_soft_placement=True,
                        device_count = {'CPU' : num_CPU,
                                        'GPU' : num_GPU}, 
                        log_device_placement=True

                       )
config.gpu_options.per_process_gpu_memory_fraction=0.35
config.gpu_options.force_gpu_compatible = True
B.set_session(tf.compat.v1.Session(config=config))

We are brave and try our first runs directly on a GPU. The statement “log_device_placement” will help us to get information about which device – CPU or GP – is actually used.

Loading and preparing MNIST data

We prepare a function which loads and prepares the MNIST data for us. The statements reflect more or less what we did with the MNIST dat when we used them for MLPs.

  
# load MNIST 
# **********
def load_Mnist():
    mnist = K.datasets.mnist
    (X_train, y_train), (X_test, y_test) = mnist.load_
data()

    #print(X_train.shape)
    #print(X_test.shape)

    # preprocess - flatten 
    len_train =  X_train.shape[0]
    len_test  =  X_test.shape[0]
    X_train = X_train.reshape(len_train, 28*28) 
    X_test  = X_test.reshape(len_test, 28*28) 

    #concatenate
    _X = np.concatenate((X_train, X_test), axis=0)
    _y = np.concatenate((y_train, y_test), axis=0)

    _dim_X = _X.shape[0]

    # 32-bit
    _X = _X.astype(np.float32)
    _y = _y.astype(np.int32)

    # normalize  
    scaler = StandardScaler()
    _X = scaler.fit_transform(_X)

    # mixing the training indices - MUST happen BEFORE encoding
    shuffled_index = np.random.permutation(_dim_X)
    _X, _y = _X[shuffled_index], _y[shuffled_index]

    # split again 
    num_test  = 10000
    num_train = _dim_X - num_test
    X_train, X_test, y_train, y_test = _X[:num_train], _X[num_train:], _y[:num_train], _y[num_train:]

    # reshape to Keras tensor requirements 
    train_imgs = X_train.reshape((num_train, 28, 28, 1))
    test_imgs  = X_test.reshape((num_test, 28, 28, 1))
    #print(train_imgs.shape)
    #print(test_imgs.shape)

    # one-hot-encoding
    train_labels = to_categorical(y_train)
    test_labels  = to_categorical(y_test)
    #print(test_labels[4])
    
    return train_imgs, test_imgs, train_labels, test_labels

if gpu:
    with tf.device("/GPU:0"):
        train_imgs, test_imgs, train_labels, test_labels= load_Mnist()
else:
    with tf.device("/CPU:0"):
        train_imgs, test_imgs, train_labels, test_labels= load_Mnist()

 
Some comments:

  • Normalization and shuffling: The “StandardScaler” is used for data normalization. I also shuffled the data to avoid any pre-ordered sequences. We know these steps already from the MLP code we built in another article series.
  • Image data in tensor form: Something, which is different from working with MLPs is that we have to fulfill some requirements regarding the form of input data. From the last article we know already that our data should have a tensor compatible form; Keras expects data from us which have a certain shape. So, no flattening of the data into a vector here as we were used to with MLPs. For images we, instead, need the width, the height of our images in terms of pixels and also the depth (here 1 for gray-scale images). In addition the data samples are to be indexed along the first tensor axis.
    This means that we need to deliver a 4-dimensional array corresponding to a TF tensor of rank 4. Keras/TF2 will do the necessary transformation from a Numpy array to a TF2 tensor automatically for us. The corresponding Numpy shape of the required array is:
    (samples, height, width, depth)
    Some people also use the term “channels” instead of “depth”. In the case of MNIST we reshape the input array – “train_imgs” to (num_train, 28, 28, 1), with “num_train” being the number of samples.
  • The use of the function “to_categorical()”, more precisely “tf.keras.utils.to_categorical()”, corresponds to a one-hot-encoding of the target data. All these concepts are well known from our study of MLPs and MNIST. Keras makes life easy regarding this point …
  • The statements “with tf.device(“/GPU:0”):” and “with tf.device(“/CPU:0”):” delegate the execution of (suitable) code on the GPU or the CPU. Note that due to the Python/Jupyter environment some code will of course also be executed on the CPU – even if you delegated execution to the GPU.

If you activate the print statements the resulting output should be:

(60000, 
28, 28)
(10000, 28, 28)
(60000, 28, 28, 1)
(10000, 28, 28, 1)
[0. 0. 0. 0. 0. 0. 0. 1. 0. 0.]

The last line proves the one-hot-encoding.

The CNN architecture – and Keras’ layer API

Now, we come to a central point: We need to build the 5 central layers of our CNN-architecture. When we build our own MLP code we used a special method to build the different weight arrays, which represented the number of nodes via the array dimensions. A simple method was sufficient as we had no major differences between layers. But with CNNs we have to work with substantially different types of layers. So, how are layers to be handled with Keras?

Well, Keras provides a full layer API with different classes for a variety of layers. You find substantial information on this API and different types of layers at
https://keras.io/api/layers/.

The first section which is interesting for our experiment is https://keras.io/api/ layers/ convolution_layers/ convolution2d/.
You do not need to read much to understand that this is exactly what we need for the “convolutional layers” of our simple CNN model. But how do we instantiate the Conv2D class such that the output works seamlessly together with other layers?

Keras makes our life easy again. All layers are to be used in a purely sequential order. (There are much more complicated layer topologies you can build with Keras! Oh, yes …). Well, you guess it: Keras offers you a model API; see:
https://keras.io/api/models/.

And there we find a class for a “sequential model” – see https://keras.io/api/ models/sequential/. This class offers us a method “add()” to add layers (and create an instance of the related layer class).

The only missing ingredient is a class for a “pooling” layer. Well, you find it in the layer API documentation, too. The following image depicts the basic structure of our CNN (see the left side of the drawing), as we designed it (see the list above):

Keras code for the Conv and pooling layers

The convolutional part of the CNN can be set up by the following commands:

Convolutional part of the CNN

# Sequential layer model of our CNN
# ***********************************

# Build the Conv part of the CNN
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

# Choose the activation function for the Conv2D layers 
conv_act_func = 1
li_conv_act_funcs = ['sigmoid', 'relu', 'elu', 'tanh']
cact = li_conv_act_funcs[conv_act_func]

# Build the Conv2D layers 
cnn = models.Sequential()
cnn.add(layers.Conv2D(32, (3,3), activation=cact, input_shape=(28,28,1)))
cnn.add(layers.MaxPooling2D((2,2)))
cnn.add(layers.Conv2D(64, (3,3), activation=cact))
cnn.add(layers.MaxPooling2D((2,2)))
cnn.add(layers.Conv2D(64, (3,3), activation=cact))

Easy, isn’t it? The nice thing about Keras is that it cares about the required tensor ranks and shapes itself; in a sequential model it evaluates the output of a already defined layer to guess the shape of the tensor data entering the next layer. Thus we have to define an “input_shape” only for data entering the first Conv2D layer!

The first Conv2D layer requires, of course, a shape for the input data. We must also tell the layer interface how many filters and “feature maps” we want to use. In our case we produce 32 maps by first Conv2D layer and 64 by the other two Conv2D layers. The (3×3)-parameter defines the filter area size to be covered by the filter kernel: 3×3 pixels. We define no “stride”, so a stride of 1 is automatically used; all 3×3 areas lie close to each other and overlap each other. These parameters result in 32 maps of size 26×26 after the first convolution. The size of the maps of the other layers are given in the layer list at the beginning of this article.

In addition you saw from the code that we chose an activation function via an index of a Python list of reasonable alternatives. You find an explanation of all the different activation functions in the literature. (See also: wikipedia: Activation function). The “sigmoid” function should be well known to you already from my other article series on a MLP.

Now, we have to care about the MLP part of the CNN. The code is simple:

MLP part of the CNN

# Build the MLP part of the CNN
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

# Choose the activation function for the hidden layers of the MLP 
mlp_h_act_func = 0
li_mlp_h_act_funcs = ['relu', 'sigmoid', 'tanh']
mhact = li_mlp_h_act_funcs[mlp_h_act_func]

# Choose the output function for the output layer of the MLP 
mlp_o_act_func = 0
li_mlp_o_act_funcs = ['softmax', 'sigmoid']
moact = li_mlp_o_act_funcs[mlp_o_act_func]

# Build the MLP layers 
cnn.add(layers.Flatten())
cnn.add(layers.Dense(70, activation=mhact))
#cnn.add(layers.Dense(30, activation=mhact))
cnn.add(layers.Dense(10, activation=moact))

This all is very straight forward (with the exception of the last statement). The “Flatten”-layer corresponds to the MLP’s inout layer. It just transforms the tensor output of the last Conv2D layer into the flat form usable for the first “Dense” layer of the MLP. The first and only “Dense layer” (MLP hidden layer) builds up connections from the flat MLP “input layer” and associates it with weights. Actually, it prepares a weight-tensor for a tensor-operation on the output data of the feeding layer. Dense means that all “nodes” of the previous layer are connected to the present layer’s own “nodes” – meaning: setting the right dimensions of the weight tensor (matrix in our case). As a first trial we work with just one hidden layer. (We shall later see that more layers will not improve accuracy.)

I choose the output function (if you will: the activation function of the output layer) as “softmax“. This gives us a probability distribution across the classification categories. Note that this is a different approach compared to what we have done so far with MLPs. I shall comment on the differences in a separate article when I find the time for it. At this point I just want to indicate that softmax combined with the “categorical cross entropy loss” is a generalized version of the combination “sigmoid” with “log loss” as we used it for our MLP.

Parameterization

The above code for creating the CNN would work. However, we want to be able to parameterize our simple CNN. So we include the above statements in a function for which we provide parameters for all layers. A quick solution is to define layer parameters as elements of a Python list – we then get one list per layer. (If you are a friend of clean code design I recommend to choose a more elaborated approach; inject just one parameter object containing all parameters in a structured way. I leave this exercise to you.)

We now combine the statements for layer construction in a function:

  
# Sequential layer model of our CNN
# ***********************************

# just for illustration - the real parameters are fed later 
# 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
li_conv_1   = [32, (3,3), 0] 
li_conv_2   = [64, (3,3), 0] 
li_conv_3   = [64, (3,3), 0] 
li_Conv     = [li_conv_1, li_conv_2, li_conv_3]
li_pool_1   = [(2,2)]
li_pool_2   = [(2,2)]
li_Pool     = [li_pool_1, li_pool_2]
li_dense_1  = [70, 0]
li_dense_2  = [10, 0]
li_MLP      = [li_dense_1, li_dense_2]
input_shape = (28,28,1)

# important !!
# ~~~~~~~~~~~
cnn = None

def build_cnn_simple(li_Conv, li_Pool, li_MLP, input_shape ):

    # activation functions to be used in Conv-layers 
    li_conv_act_funcs = ['relu', 'sigmoid', 'elu', 'tanh']
    # activation functions to be used in MLP hidden layers  
    li_mlp_h_act_funcs = ['relu', 'sigmoid', 'tanh']
    # activation functions to be used in MLP output layers  
    li_mlp_o_act_funcs = ['softmax', 'sigmoid']

    
    # Build the Conv part of the CNN
    # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    num_conv_layers = len(li_Conv)
    num_pool_layers = len(li_Pool)
    if num_pool_layers != num_conv_layers - 1: 
        print("\nNumber of pool layers does not fit to number of Conv-layers")
        sys.exit()
    rg_il = range(num_conv_layers)

    # Define a sequential model 
    cnn = models.Sequential()

    for il in rg_il:
        # add the convolutional layer 
        num_filters = li_Conv[il][0]
        t_fkern_size = li_Conv[il][1]
        cact        = li_conv_act_funcs[li_Conv[il][2]]
        if il==0:
            cnn.add(layers.Conv2D(num_filters, t_fkern_size, activation=cact, input_shape=input_shape))
        else:
            cnn.add(layers.Conv2D(num_filters, t_fkern_size, activation=cact))
        
        # add the pooling layer 
        if il < num_pool_layers:
            t_pkern_size = li_Pool[il][0]
            cnn.add(layers.MaxPooling2D(t_pkern_size))
            

    # Build the MLP part of the CNN
    # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    num_mlp_layers = len(li_MLP)
    rg_im = range(num_mlp_layers)

    cnn.add(layers.Flatten())

    for im in rg_im:
        # add the dense layer 
        n_nodes = li_MLP[im][0]
        if im < num_mlp_layers - 1:  
            m_act   =  li_mlp_h_act_funcs[li_MLP[im][1]]
        else: 
            m_act   =  li_mlp_o_act_funcs[li_MLP[im][1]]
        cnn.add(layers.Dense(n_nodes, activation=m_act))

    return cnn 

 

We return the model “cnn” to be able to use it afterwards.

How many parameters does our CNN have?

The layers contribute with the following numbers of weight parameters:

  • Layer 1: (32 x (3×3)) + 32 = 320
  • Layer 3: 32 x 64 x (3×3) + 64 = 18496
  • Layer 5: 64 x 64 x (3×3) + 64 = 36928
  • MLP : (576 + 1) x 70 + (70 + 1) x 10 = 41100

Making a total of 96844 weight parameters. Our standard MLP discussed in another article series had (784+1) x 70 + (70 + 1) x 30 + (30 +1 ) x 10 = 57390 weights. So, our CNN is bigger and the CPU time to follow and calculate all the partial derivatives will be significantly higher. So, we should definitely expect some better classification data, shouldn’t we?

Compilation

Now comes a thing which is necessary for models: We have not yet defined the loss function and the optimizer or a learning rate. For the latter Keras can choose a proper value itself – as soon as it knows the loss function. But we should give it a reasonable loss function and a suitable optimizer for gradient descent. This is the main purpose of the “compile()“-function.

cnn.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])

Although TF2 can already analyze the graph of tensor operations for partial derivatives, it cannot guess the beginning of the chain rule sequence.

As we have multiple categories “categorial_crossentropy” is a good choice for the loss function. We should also define which optimized gradient descent method is used; we choose “rmsprop” – as this method works well in most cases. A nice introduction is given here: towardsdatascience: understanding-rmsprop-faster-neural-network-learning-62e116fcf29a. But see the books mentioned in the last article on “rmsprop”, too.

Regarding the use of different metrics for different tasks see machinelearningmastery.com / custom-metrics-deep-learning-keras-python/. In case of a classification problem, we are interested in the categorical “accuracy”. A metric can be monitored during training and will be recorded (besides aother data). We can use it for plotting information on the training process (a topic of the next article).

Training

Training is done by a function model.fit() – here: cnn.fit(). This function accepts a variety of parameters explained here: https://keras.io/ api/ models/ model_training_apis/ #fit-method.

We now can combine compilation and training in one function:

# Training 
def train( cnn, build=False, train_imgs, train_labels, reset, epochs, batch_size, optimizer, loss, metrics,
           li_Conv, li_Poo, li_MLP, input_shape ):
    if build:
        cnn = build_cnn_simple( li_Conv, li_Pool, li_MLP, input_shape)
        cnn.compile(optimizer=optimizer, loss=loss, metrics=metrics)        
        cnn.save_weights('cnn_weights.h5') # save the initial weights 
    # reset weights
    if reset and not build:
        cnn.load_weights('cnn_weights.h5')
    start_t = time.perf_counter()
    cnn.fit(train_imgs, train_labels, epochs=epochs, batch_size=batch_size, verbose=1, shuffle=True) 
    end_t = time.perf_counter()
    fit_t = end_t - start_t
    return cnn, fit_t  # we return cnn to be able to use it by other functions

Note that we save the initial weights to be able to load them again for a new training – otherwise Keras saves the weights as other model data after training and continues with the last weights found. The latter can be reasonable if you want to continue training in defined steps. However, in our forthcoming tests we repeat the training from scratch.

Keras offers a “save”-model and methods to transfer data of a CNN model to files (in two specific formats). For saving weights the given lines are sufficient. Hint: When I specify no path to the file “cnn_weights.h5” the data are – at least in my virtual Python environment – saved in the directory where the notebook is located.

First test

In a further Jupyter cell we place the following code for a test run:

  
# Perform a training run 
# ********************
build = False     
if cnn == None:
    build = True
batch_size=64
epochs=5
reset = True # we want training to start again with the initial weights

optimizer='rmsprop' 
loss='categorical_crossentropy'
metrics=['accuracy']

li_conv_1   = [32, (3,3), 0] 
li_conv_2   = [64, (3,3), 0] 
li_conv_3   = [64, (3,3), 0] 
li_Conv     = [li_conv_1, li_conv_2, li_conv_3]
li_pool_1   = [(2,2)]
li_pool_2   = [(2,2)]
li_Pool     = [li_pool_1, li_pool_2]
li_dense_1  = [70, 0]
li_dense_2  = [10, 0]
li_MLP      = [li_dense_1, li_dense_2]
input_shape = (28,28,1)

try: 
    if gpu:
        with tf.device("/GPU:0"):
            cnn, fit_time = train( cnn, build, train_imgs, train_
labels, 
                                   reset, epochs, batch_size, optimizer, loss, metrics, 
                                   li_Conv, li_Pool, li_MLP, input_shape)
        print('Time_GPU: ', fit_time)  
    else:
        with tf.device("/CPU:0"):
            cnn, fit_time = train( cnn, build, train_imgs, train_labels, 
                                   reset, epochs, batch_size, optimizer, loss, metrics, 
                                   li_Conv, li_Pool, li_MLP, input_shape)
        print('Time_CPU: ', fit_time)  
except SystemExit:
    print("stopped due to exception")

You recognize the parameterization of our train()-function. What results do we get ?

Epoch 1/5
60000/60000 [==============================] - 4s 69us/step - loss: 0.1551 - accuracy: 0.9520
Epoch 2/5
60000/60000 [==============================] - 4s 69us/step - loss: 0.0438 - accuracy: 0.9868
Epoch 3/5
60000/60000 [==============================] - 4s 68us/step - loss: 0.0305 - accuracy: 0.9907
Epoch 4/5
60000/60000 [==============================] - 4s 69us/step - loss: 0.0227 - accuracy: 0.9931
Epoch 5/5
60000/60000 [==============================] - 4s 69us/step - loss: 0.0184 - accuracy: 0.9948
Time_GPU:  20.610678611003095

 

And a successive check on the test data gives us:

We can ask Keras also for a description of the model:

Accuracy at the 99% level

We got an accuracy on the test data of 99%! With 5 epochs in 20 seconds – on my old GPU.
This leaves us a very good impression – on first sight …

Conclusion

We saw today that it is easy to set up a CNN. We used a simple MLP to solve the problem of classification; the data to its input layer are provided by the output of the last convolutional layer. The tensor there has just to be “flattened”.

The level of accuracy reached is impressing. Well, its also a bit frustrating when we think about the efforts we put into our MLP, but we also get a sense for the power and capabilities of CNNs.

In the next post
A simple CNN for the MNIST dataset – III – inclusion of a learning-rate scheduler, momentum and a L2-regularizer
we will care a bit about plotting. We at least want to see the same accuracy and loss data which we used to plot at the end of our MLP tests.

 

MLP, Numpy, TF2 – performance issues – Step III – a correction to BW propagation

In the last articles of this series

MLP, Numpy, TF2 – performance issues – Step II – bias neurons, F- or C- contiguous arrays and performance
MLP, Numpy, TF2 – performance issues – Step I – float32, reduction of back propagation

we looked at the FW-propagation of the MLP code which I discussed in another article series. We found that the treatment of bias neurons in the input layer was technically inefficient due to a collision of C- and F-contiguous arrays. By circumventing the problem we could accelerate the FW-propagation of big batches (as the complete training or test data set) by more than a factor of 2.

In this article I want to turn to the BW propagation and do some analysis regarding CPU consumption there. We will find a simple (and stupid) calculation step there which we shall replace. This will give us another 15% to 22% performance improvement in comparison to what we have reached in the last article for MNIST data:

  • 9.6 secs for 35 epochs and a batch-size of 500
  • and 8.7 secs for a batch-size of 20000.

Present CPU time relation between the FW- and the BW-propagation

The central training of mini-batches is performed by the method “_handle_mini_batch()”.

#
    ''' -- Method to deal with a batch -- '''
    def _handle_mini_batch (self, num_batch = 0, num_epoch = 0, b_print_y_vals = False, b_print = False, b_keep_bw_matrices = True):
        ''' .... '''
        # Layer-related lists to be filled with 2-dim Numpy matrices during FW propagation
        # ********************************************************************************
        li_Z_in_layer  = [None] * self._n_total_layers # List of matrices with z-input values for each layer; filled during FW-propagation
        li_A_out_layer = li_Z_in_layer.copy()          # List of matrices with results of activation/output-functions for each layer; filled during FW-propagation
        li_delta_out   = li_Z_in_layer.copy()          # Matrix with out_delta-values at the outermost layer 
        li_delta_layer = li_Z_in_layer.copy()          # List of the matrices for the BW propagated delta values 
        li_D_layer     = li_Z_in_layer.copy()          # List of the derivative matrices D containing partial derivatives of the activation/ouput functions 
        li_grad_layer  = li_Z_in_layer.copy()          # List of the matrices with gradient values for weight corrections
        
        # Major steps for the mini-batch during one epoch iteration 
        # **********************************************************
        
        #ts=time.perf_counter()
        # Step 0: List of indices for data records in the present mini-batch
        # ******
        ay_idx_batch = self._ay_mini_batches[num_batch]
        
        # Step 1: Special preparation of the Z-input to the MLP's input Layer L0
        # ******
        # ts=time.perf_counter()
        # slicing 
        li_Z_in_layer[0] = self._X_train[ay_idx_batch] # numpy arrays can be indexed by an array of integers
        
        # transposition 
        #~~~~~~~~~~~~~~
        li_Z_in_layer[0] = li_Z_in_layer[0].T
        #te=time.perf_counter(); t_batch = te - ts;
        #print("\nti - transposed inputbatch =", t_batch)
        
        # Step 2: Call forward propagation method for the present mini-batch of training records
        # *******
n        #tsa = time.perf_counter() 
        self._fw_propagation(li_Z_in = li_Z_in_layer, li_A_out = li_A_out_layer) 
        #tea = time.perf_counter(); ta = tea - tsa;  print("ta - FW-propagation", "%10.8f"%ta)
        
        # Step 3: Cost calculation for the mini-batch 
        # ********
        #tsb = time.perf_counter() 
        ay_y_enc = self._ay_onehot[:, ay_idx_batch]
        ay_ANN_out = li_A_out_layer[self._n_total_layers-1]
        total_costs_batch, rel_reg_contrib = self._calculate_loss_for_batch(ay_y_enc, ay_ANN_out, b_print = False)
        # we add the present cost value to the numpy array 
        self._ay_costs[num_epoch, num_batch]            = total_costs_batch
        self._ay_reg_cost_contrib[num_epoch, num_batch] = rel_reg_contrib
        #teb = time.perf_counter(); tb = teb - tsb; print("tb - cost calculation", "%10.8f"%tb)
        
        
        # Step 4: Avg-error for later plotting 
        # ********
        #tsc = time.perf_counter() 
        # mean "error" values - averaged over all nodes at outermost layer and all data sets of a mini-batch 
        ay_theta_out = ay_y_enc - ay_ANN_out
        ay_theta_avg = np.average(np.abs(ay_theta_out)) 
        self._ay_theta[num_epoch, num_batch] = ay_theta_avg 
        #tec = time.perf_counter(); tc = tec - tsc; print("tc - error", "%10.8f"%tc)
        
        
        # Step 5: Perform gradient calculation via back propagation of errors
        # ******* 
        #tsd = time.perf_counter() 
        self._bw_propagation( ay_y_enc = ay_y_enc, 
                              li_Z_in = li_Z_in_layer, 
                              li_A_out = li_A_out_layer, 
                              li_delta_out = li_delta_out, 
                              li_delta = li_delta_layer,
                              li_D = li_D_layer, 
                              li_grad = li_grad_layer, 
                              b_print = b_print,
                              b_internal_timing = False 
                              ) 
        #ted = time.perf_counter(); td = ted - tsd; print("td - BW propagation", "%10.8f"%td)
        
        # Step 7: Adjustment of weights  
        # *******        
        #tsf = time.perf_counter() 
        rg_layer=range(0, self._n_total_layers -1)
        for N in rg_layer:
            delta_w_N = self._learn_rate * li_grad_layer[N]
            self._li_w[N] -= ( delta_w_N + (self._mom_rate * self._li_mom[N]) )
            
            # save momentum
            self._li_mom[N] = delta_w_N
        #tef = time.perf_counter(); tf = tef - tsf; print("tf - weight correction", "%10.8f"%tf)
        
        return None

 

I took some time measurements there:

ti - transposed inputbatch = 0.0001785
ta - FW-propagation 0.00080975
tb - cost calculation 0.00030705
tc - error 0.00016182
td - BW propagation 0.00112558
tf - weight correction 0.00020079

ti - transposed inputbatch = 0.00018144
ta - FW-propagation 0.00082022
tb - cost calculation 0.00031284
tc - error 0.00016652
td - BW propagation 0.00106464
tf - weight correction 0.00019576

You see that the FW-propagation is a bit faster than the BW-propagation. This is a bit strange as the FW-propagation is dominated meanwhile by a really expensive operation which we cannot accelerate (without choosing a new activation function): The calculation of the sigmoid value for the inputs at layer L1.

So let us look into the BW-propagation; the code for it is momentarily:

    ''' -- Method to handle error BW propagation for a mini-batch --'''
    def _bw_propagation(self, 
                        ay_y_enc, li_Z_in, li_A_out, 
                        li_delta_out, li_delta, li_D, li_
grad, 
                        b_print = True, b_internal_timing = False):
        
        # List initialization: All parameter lists or arrays are filled or to be filled by layer operations 
        # Note: the lists li_Z_in, li_A_out were already filled by _fw_propagation() for the present batch 
        
        # Initiate BW propagation - provide delta-matrices for outermost layer
        # *********************** 
        tsa = time.perf_counter() 
        # Input Z at outermost layer E  (4 layers -> layer 3)
        ay_Z_E = li_Z_in[self._n_total_layers-1]
        # Output A at outermost layer E (was calculated by output function)
        ay_A_E = li_A_out[self._n_total_layers-1]
        
        # Calculate D-matrix (derivative of output function) at outmost the layer - presently only D_sigmoid 
        ay_D_E = self._calculate_D_E(ay_Z_E=ay_Z_E, b_print=b_print )
        #ay_D_E = ay_A_E * (1.0 - ay_A_E)

        # Get the 2 delta matrices for the outermost layer (only layer E has 2 delta-matrices)
        ay_delta_E, ay_delta_out_E = self._calculate_delta_E(ay_y_enc=ay_y_enc, ay_A_E=ay_A_E, ay_D_E=ay_D_E, b_print=b_print) 
        
        # add the matrices to their lists ; li_delta_out gets only one element 
        idxE = self._n_total_layers - 1
        li_delta_out[idxE] = ay_delta_out_E # this happens only once
        li_delta[idxE]     = ay_delta_E
        li_D[idxE]         = ay_D_E
        li_grad[idxE]      = None    # On the outermost layer there is no gradient ! 
        
        tea = time.perf_counter(); ta = tea - tsa; print("\nta-bp", "%10.8f"%ta)
        
        # Loop over all layers in reverse direction 
        # ******************************************
        # index range of target layers N in BW direction (starting with E-1 => 4 layers -> layer 2))
        range_N_bw_layer = reversed(range(0, self._n_total_layers-1))   # must be -1 as the last element is not taken 
        
        # loop over layers 
        tsb = time.perf_counter() 
        for N in range_N_bw_layer:
            
            # Back Propagation operations between layers N+1 and N 
            # *******************************************************
            # this method handles the special treatment of bias nodes in Z_in, too
            tsib = time.perf_counter() 
            ay_delta_N, ay_D_N, ay_grad_N = self._bw_prop_Np1_to_N( N=N, li_Z_in=li_Z_in, li_A_out=li_A_out, li_delta=li_delta, b_print=False )
            teib = time.perf_counter(); tib = teib - tsib; print("N = ", N, " tib-bp", "%10.8f"%tib)
            
            # add matrices to their lists 
            #tsic = time.perf_counter() 
            li_delta[N] = ay_delta_N
            li_D[N]     = ay_D_N
            li_grad[N]= ay_grad_N
            #teic = time.perf_counter(); tic = teic - tsic; print("\nN = ", N, " tic = ", "%10.8f"%tic)
        teb = time.perf_counter(); tb = teb - tsb; print("tb-bp", "%10.8f"%tb)
       
        return

 

Typical timing results are:

ta-bp 0.00007112
N =  2  tib-bp 0.00025399
N =  1  tib-bp 0.00051683
N =  0  tib-bp 0.00035941
tb-bp 0.00126436

ta-bp 0.00007492
N =  2  tib-bp 0.00027644
N =  1  tib-bp 0.00090043
N =  0  tib-bp 0.00036728
tb-bp 0.00168378

We see that the CPU consumption of “_bw_prop_Np1_to_N()” should be analyzed in detail. It is relatively time consuming at every layer, but especially at layer L1. (The list adds are insignificant.)
What does this method presently look like?

    ''' -- Method to calculate the BW-propagated delta-matrix and the gradient matrix to/for layer N '''
    def _bw_prop_Np1_to_N(self, N, li_Z_in, li_A_out, li_delta, b_print=False):
        '''
        BW-
error-propagation between layer N+1 and N 
        Version 1.5 - partially accelerated 

        Inputs: 
            li_Z_in:  List of input Z-matrices on all layers - values were calculated during FW-propagation
            li_A_out: List of output A-matrices - values were calculated during FW-propagation
            li_delta: List of delta-matrices - values for outermost ölayer E to layer N+1 should exist 
        
        Returns: 
            ay_delta_N - delta-matrix of layer N (required in subsequent steps)
            ay_D_N     - derivative matrix for the activation function on layer N 
            ay_grad_N  - matrix with gradient elements of the cost fnction with respect to the weights on layer N 
        '''
        
        # Prepare required quantities - and add bias neuron to ay_Z_in 
        # ****************************
        
        # Weight matrix meddling between layers N and N+1 
        ay_W_N = self._li_w[N]

        # delta-matrix of layer N+1
        ay_delta_Np1 = li_delta[N+1]

        # fetch output value saved during FW propagation 
        ay_A_N = li_A_out[N]

        # Optimization V1.5 ! 
        if N > 0: 
            
            #ts=time.perf_counter()
            ay_Z_N = li_Z_in[N]
            # !!! Add intermediate row (for bias) to Z_N !!!
            ay_Z_N = self._add_bias_neuron_to_layer(ay_Z_N, 'row')
            #te=time.perf_counter(); t1 = te - ts; print("\nBW t1 = ", t1, " N = ", N) 
        
            # Derivative matrix for the activation function (with extra bias node row)
            # ********************
            #    can only be calculated now as we need the z-values
            #ts=time.perf_counter()
            ay_D_N = self._calculate_D_N(ay_Z_N)
            #te=time.perf_counter(); t2 = te - ts; print("\nBW t2 = ", t2, " N = ", N) 
            
            # Propagate delta
            # **************

            # intermediate delta 
            # ~~~~~~~~~~~~~~~~~~
            #ts=time.perf_counter()
            ay_delta_w_N = ay_W_N.T.dot(ay_delta_Np1)
            #te=time.perf_counter(); t3 = te - ts; print("\nBW t3 = ", t3) 
            
            # final delta 
            # ~~~~~~~~~~~
            #ts=time.perf_counter()
            ay_delta_N = ay_delta_w_N * ay_D_N
            
            # Orig reduce dimension again
            # **************************** 
            ay_delta_N = ay_delta_N[1:, :]
            #te=time.perf_counter(); t4 = te - ts; print("\nBW t4 = ", t4) 
            
        else: 
            ay_delta_N = None
            ay_D_N = None
        
        # Calculate gradient
        # ********************
        #ts=time.perf_counter()
        ay_grad_N = np.dot(ay_delta_Np1, ay_A_N.T)
        #te=time.perf_counter(); t5 = te - ts; print("\nBW t5 = ", t5) 
        
        # regularize gradient (!!!! without adding bias nodes in the L1, L2 sums) 
        #ts=time.perf_counter()
        if self._lambda2_reg > 0: 
            ay_grad_N[:, 1:] += self._li_w[N][:, 1:] * self._lambda2_reg 
        if self._lambda1_reg > 0: 
            ay_grad_N[:, 1:] += np.sign(self._li_w[N][:, 1:]) * self._lambda1_reg 
        #te=time.perf_counter(); t6 = te - ts; print("\nBW t6 = ", t6) 
        
        return ay_delta_N, ay_D_N, ay_grad_N

 
Timing data for a batch-size of 500 are:

N =  2
BW t1 =  0.0001169009999557602  N =  2
BW t2 =  0.00035331499998392246  N =  2
BW t3 =  0.00018078099992635543
BW t4 =  0.00010234199999104021
BW t5 =  9.928200006470433e-05
BW t6 =  2.4267000071631628e-05
N =  2  tib-bp 0.00124414

N =  1
BW t1 =  0.0004323499999827618  N =  1
BW t2 =  0.
000781415999881574  N =  1
BW t3 =  4.2077999978573644e-05
BW t4 =  0.00022921000004316738
BW t5 =  9.376399998473062e-05
BW t6 =  0.00012183700005152787
N =  1  tib-bp 0.00216281

N =  0
BW t5 =  0.0004289769999559212
BW t6 =  0.00015404999999191205
N =  0  tib-bp 0.00075249
....
N =  2
BW t1 =  0.00012802800006284087  N =  2
BW t2 =  0.00034988200013685855  N =  2
BW t3 =  0.0001854429999639251
BW t4 =  0.00010359299994888715
BW t5 =  0.00010210400000687514
BW t6 =  2.4010999823076418e-05
N =  2  tib-bp 0.00125854

N =  1
BW t1 =  0.0004407169999467442  N =  1
BW t2 =  0.0007845899999665562  N =  1
BW t3 =  0.00025684100000944454
BW t4 =  0.00012409999999363208
BW t5 =  0.00010345399982725212
BW t6 =  0.00012994100006835652
N =  1  tib-bp 0.00221321

N =  0
BW t5 =  0.00044504700008474174
BW t6 =  0.00016473000005134963
N =  0  tib-bp 0.00071442

....
N =  2
BW t1 =  0.000292730999944979  N =  2
BW t2 =  0.001102525000078458  N =  2
BW t3 =  2.9429999813146424e-05
BW t4 =  8.547999868824263e-06
BW t5 =  3.554099998837046e-05
BW t6 =  2.5041999833774753e-05
N =  2  tib-bp 0.00178565

N =  1
BW t1 =  3.143399999316898e-05  N =  1
BW t2 =  0.0006720640001276479  N =  1
BW t3 =  5.4785999964224175e-05
BW t4 =  9.756200006449944e-05
BW t5 =  0.0001605449999715347
BW t6 =  1.8391000139672542e-05
N =  1  tib-bp 0.00147566

N =  0
BW t5 =  0.0003641810001226986
BW t6 =  6.338999992294703e-05
N =  0  tib-bp 0.00046542

 
It seems that we should care about t1, t2, t3 for hidden layers and maybe about t5 at layers L1/L0.

However, for a batch-size of 15000 things look a bit different:

N =  2
BW t1 =  0.0005776280000304723  N =  2
BW t2 =  0.004995969999981753  N =  2
BW t3 =  0.0003165199999557444
BW t4 =  0.0005244750000201748
BW t5 =  0.000518499999998312
BW t6 =  2.2458999978880456e-05
N =  2  tib-bp 0.00736144

N =  1
BW t1 =  0.0010120430000029046  N =  1
BW t2 =  0.010797029000002567  N =  1
BW t3 =  0.0005006920000028003
BW t4 =  0.0008704929999794331
BW t5 =  0.0010805200000163495
BW t6 =  3.0326000000968634e-05
N =  1  tib-bp 0.01463436

N =  0
BW t5 =  0.006987539000022025
BW t6 =  0.00023552499999368592
N =  0  tib-bp 0.00730959


N =  2
BW t1 =  0.0006299790000525718  N =  2
BW t2 =  0.005081416999985322  N =  2
BW t3 =  0.00018547400003399162
BW t4 =  0.0005970070000103078
BW t5 =  0.000564008000026206
BW t6 =  2.3311000006742688e-05
N =  2  tib-bp 0.00737899

N =  1
BW t1 =  0.0009376909999900818  N =  1
BW t2 =  0.010650266999959968  N =  1
BW t3 =  0.0005232729999988806
BW t4 =  0.0009100700000317374
BW t5 =  0.0011237720000281115
BW t6 =  0.00016643800000792908
N =  1  tib-bp 0.01466144

N =  0
BW t5 =  0.006987463000029948
BW t6 =  0.00023978600000873485
N =  0  tib-bp 0.00734308

 
For big batch-sizes “t2” dominates everything. It seems that we have found another code area which causes the trouble with big batch-sizes which we already observed before!

What operations do the different CPU times stand for?

To keep an overview without looking into the code again, I briefly summarize which operations cause which of the measured time differences:

  • t1” – which contributes for small batch-sizes stands for adding a bias neuron to the input data Z_in at each layer.
  • t2” – which is by far dominant for big batch sizes stands for calculating the derivative of the output/activation function (in our case of the sigmoid function) at the various layers.
  • t3” – which contributes at
    some layers stands for a dot()-matrix multiplication with the transposed weight-matrix,
  • t4” – covers an element-wise matrix-multiplication,
  • t5” – contributes at the BW-transition from layer L1 to L0 and covers the matrix multiplication there (including the full output matrix with the bias neurons at L0)

Use the output values calculated at each layer during FW-propagation!

Why does the calculation of the derivative of the sigmoid function take so much time? Answer: Because I coded it stupidly! Just look at it:

    ''' -- Method to calculate the matrix with the derivative values of the output function at outermost layer '''
    def _calculate_D_N(self, ay_Z_N, b_print= False):
        '''
        This method calculates and returns the D-matrix for the outermost layer
        The D matrix contains derivatives of the output function with respect to local input "z_j" at outermost nodes. 
        
        Returns
        ------
        ay_D_E:    Matrix with derivative values of the output function 
                   with respect to local z_j valus at the nodes of the outermost layer E
        Note: This is a 2-dim matrix over layer nodes and training samples of the mini-batch
        '''
        if self._my_out_func == 'sigmoid':
            ay_D_E = self._D_sigmoid(ay_Z = ay_Z_N)
        
        else:
            print("The derivative for output function " + self._my_out_func + " is not known yet!" )
            sys.exit()
        
        return ay_D_E

    ''' -- method for the derivative of the sigmoid function-- '''
    def _D_sigmoid(self, ay_Z):
        ''' 
        Derivative of sigmoid function with respect to Z-values 
        - works via expit element-wise on matrices
        Input:  Z - Matrix with Input values for the activation function Phi() = sigmoid() 
        Output: D - Matrix with derivative values 
        '''
        S_Z = self._sigmoid(ay_Z)
        return S_Z * (1.0 - S_Z)

 
We first call an intermediate function which then directs us to the right function for a chosen activation function. Well meant: So far, we use only the sigmoid function, but it could e.g. also be the relu() or tanh()-function. So, we did what we did for the sake of generalization. But we did it badly because of two reasons:

  • We did not keep up a function call pattern which we introduced in the FW-propagation.
  • The calculation of the derivative is inefficient.

The first point is a minor one: During FW-propagation we called the right (!) activation function, i.e. the one we choose by input parameters to our ANN-object, by an indirect call. Why not do it the same way here? We would avoid an intermediate function call and keep up a pattern. Actually, we prepared the necessary definitions already in the __init__()-function.

The second point is relevant for performance: The derivative function produces the correct results for a given “ay_Z”, but this is totally inefficient in our BW-situation. The code repeats a really expensive operation which we have already performed during FW-propagation: calling sigmoid(ay_Z) to get “A_out”-values per layer then. We even put the A_out-values [=sigmoid(ay_Z_in)] per layer and batch (!) with some foresight into a list in “li_A_out[]” at that point of the code (see the FW-propagation code discussed in the last article).

So, of course, we should use these “A_out”-values now in the BW-steps! No further comment …. you see what we need to do.

Hint: Actually, also other activation functions “act(Z)” like e.g. the “tanh()”-function have derivatives which depend on on “A=act(
Z)”, only. So, we should provide Z and A via an interface to the derivative function and let the respective functions take what it needs.
But, my insight into my own dumbness gets worse.

Eliminate the bias neuron operation!

Why did we need a bias-neuron operation? Answer: We do not need it! It was only introduced due to insufficient cleverness. In the article

A simple Python program for an ANN to cover the MNIST dataset – VII – EBP related topics and obstacles

I have already indicated that we use the function for adding a row of bias-neurons again only to compensate one deficit: The matrix of the derivative values did not fit the shape of the weight matrix for the required element-wise operations. However, I also said: There probably is an alternative.

Well, let me make a long story short: The steps behind t1 up to t4 to calculate “ay_delta_N” for the present layer L_N (with N>=1) can be compressed into two relatively simple lines:

ay_delta_w_N = ay_W_N.T.dot(ay_delta_Np1)
ay_delta_N = ay_delta_w_N[1:,:] * ay_A_N[1:,:] * (1.0 – ay_A_N[1:,:]); ay_D_N = None;

No bias back and forth corrections! Instead we use simple slicing to compensate for our weight matrices with a shape covering an extra row of bias node output. No Z-based derivative calculation; no sigmoid(Z)-call. The last statement is only required to support the present output interface. Think it through in detail; the shortcut does not cause any harm.

Code change for tests

Before we bring the code into a new consolidated form with re-coded methods let us see what we gain by just changing the code to the two lines given above in terms of CPU time and performance. Our function “_bw_prop_Np1_to_N()” then gets reduced to the following lines:

    ''' -- Method to calculate the BW-propagated delta-matrix and the gradient matrix to/for layer N '''
    def _bw_prop_Np1_to_N(self, N, li_Z_in, li_A_out, li_delta, b_print=False):
        
        # Weight matrix meddling between layers N and N+1 
        ay_W_N = self._li_w[N]
        ay_delta_Np1 = li_delta[N+1]

        # fetch output value saved during FW propagation 
        ay_A_N = li_A_out[N]

        # Optimization from previous version  
        if N > 0: 
            #ts=time.perf_counter()
            ay_Z_N = li_Z_in[N]
            
            # Propagate delta
            # ~~~~~~~~~~~~~~~~~
            ay_delta_w_N = ay_W_N.T.dot(ay_delta_Np1)
            ay_delta_N = ay_delta_w_N[1:,:] * ay_A_N[1:,:] * (1.0 - ay_A_N[1:,:])
            ay_D_N = None; 
            
        else: 
            ay_delta_N = None
            ay_D_N = None
        
        # Calculate gradient
        # ********************
        ay_grad_N = np.dot(ay_delta_Np1, ay_A_N.T)
        
        if self._lambda2_reg > 0: 
            ay_grad_N[:, 1:] += self._li_w[N][:, 1:] * self._lambda2_reg 
        if self._lambda1_reg > 0: 
            ay_grad_N[:, 1:] += np.sign(self._li_w[N][:, 1:]) * self._lambda1_reg 
        
        return ay_delta_N, ay_D_N, ay_grad_N

 

Performance gain

What run times do we get with this setting? We perform our typical test runs over 35 epochs – but this time for two different batch-sizes:

Batch-size = 500

 
------------------
Starting epoch 35

Time_CPU for epoch 35 0.2169024469985743
Total CPU-time:  7.52385053600301

learning rate =  0.0009994051838157095

total costs of training set   =  -1.0
rel. reg. contrib. to total costs =  -1.0

total costs 
of last mini_batch   =  65.43618
rel. reg. contrib. to batch costs =  0.12302863

mean abs weight at L0 :  -10.0
mean abs weight at L1 :  -10.0
mean abs weight at L2 :  -10.0

avg total error of last mini_batch =  0.00758
presently batch averaged accuracy   =  0.99272

-------------------
Total training Time_CPU:  7.5257336139984545

Not bad! We became faster by around 2 secs compared to the results of the last article! This is close to an improvement of 20%.

But what about big batch sizes? Here is the result for a relatively big batch size:

Batch-size = 20000

------------------
Starting epoch 35

Time_CPU for epoch 35 0.2019189490019926
Total CPU-time:  6.716679593999288

learning rate =  9.994051838157101e-05

total costs of training set   =  -1.0
rel. reg. contrib. to total costs =  -1.0

total costs of last mini_batch   =  13028.141
rel. reg. contrib. to batch costs =  0.00021923862

mean abs weight at L0 :  -10.0
mean abs weight at L1 :  -10.0
mean abs weight at L2 :  -10.0

avg total error of last mini_batch =  0.04389
presently batch averaged accuracy   =  0.95602

-------------------
Total training Time_CPU:  6.716954112998792

Again an acceleration by roughly 2 secs – corresponding to an improvement of 22%!

In both cases I took the best result out of three runs.

Conclusion

Enough for today! We have done a major step with regard to performance optimization also in the BW-propagation. It remains to re-code the derivative calculation in form which uses indirect function calls to remain flexible. I shall give you the code in the next article.

We learned today is that we, of course, should reuse the results of the FW-propagation and that it is indeed a good investment to save the output data per layer in some Python list or other suitable structures during FW-propagation. We also saw again that a sufficiently efficient bias neuron treatment can be achieved by a more efficient solution than provisioned so far.

All in all we have meanwhile gained more than a factor of 6.5 in performance since we started with optimization. Our new standard values are 7.3 secs and 6.8 secs for 35 epochs on MNIST data and batch sizes of 500 and 20000, respectively.

We have reached the order of what Keras and TF2 can deliver on a CPU for big batch sizes. For small batch sizes we are already faster. This indicates that we have done no bad job so far …

In the next article we shall look a bit at the matrix operations and evaluate further optimization options.

A simple Python program for an ANN to cover the MNIST dataset – XIII – the impact of regularization

I continue with my growing series on a Multilayer perceptron and the MNIST dataset.

A simple Python program for an ANN to cover the MNIST dataset – XII – accuracy evolution, learning rate, normalization
A simple Python program for an ANN to cover the MNIST dataset – XI – confusion matrix
A simple Python program for an ANN to cover the MNIST dataset – X – mini-batch-shuffling and some more tests
A simple Python program for an ANN to cover the MNIST dataset – IX – First Tests
A simple Python program for an ANN to cover the MNIST dataset – VIII – coding Error Backward Propagation
A simple Python program for an ANN to cover the MNIST dataset – VII – EBP related topics and obstacles
A simple Python program for an ANN to cover the MNIST dataset – VI – the math behind the „error back-propagation“
A simple Python program for an ANN to cover the MNIST dataset – V – coding the loss function
A simple Python program for an ANN to cover the MNIST dataset – IV – the concept of a cost or loss function
A simple Python program for an ANN to cover the MNIST dataset – III – forward propagation
A simple Python program for an ANN to cover the MNIST dataset – II – initial random weight values
A simple Python program for an ANN to cover the MNIST dataset – I – a starting point

In the last article of the series we made some interesting experiences with the variation of the “leaning rate”. We also saw that a reasonable range for initial weight values should be chosen.

Even more fascinating was, however, the impact of a normalization of the input data on a smooth and fast gradient descent. We drew the conclusion that normalization is of major importance when we use the sigmoid function as the MLP’s activation function – especially for nodes in the first hidden layer and for input data which are on average relatively big. The reason for our concern were saturation effects of the sigmoid functions and other functions with a similar variation with their argument. In the meantime I have tried to make the importance of normalization even more plausible with the help of a a very minimalistic perceptron for which we can analyze saturation effects a bit more in depth; you get to the related article series via the following link:

A single neuron perceptron with sigmoid activation function – III – two ways of applying Normalizer

There we also have a look at other normalizers or feature scalers.

But back to our series on a multi-layer perceptron. You may have have asked yourself in the meantime: Why did he not check the impact of the regularization? Indeed: We kept the parameter Lambda2 for the quadratic regularization term constant in all experiments so far: Lambda2 = 0.2. So, the question about the impact of regularization e.g. on accuracy is a good one.

How big is the regularization term and how does it evolve during gradient decent training?

I add even one more question: How big is the relative contribution of the regularization term to the total loss or cost function? In our Python program for a MLP model we included a so called quadratic Ridge term:

Lambda2 * 0.5 * SUM[all weights**2], where bias nodes are excluded from the sum.

From various books on Machine Learning [ML] you just learn to choose the factor Lambda2 in the range between 0.01 and 0.1. But how big is the resulting term actually in comparison to the standard cost term, then, and how does the ratio between both terms evolve during gradient descent? What factors influence this ratio?

As we follow a training strategy based on mini-batches the regularization contribution was and is added up to the costs of each mini-batch. So its relative importance varies of course with the size of the mini-batches! Other factors which may also be of some importance – at least during the first epochs – could be the total number of weights in our network and the range of initial weight values.

Regarding the evolution during a converging gradient descent we know already that the total costs go down on the path to a cost minimum – whilst the weight values reach a stable level. So there is a (non-linear!) competition between the regularization term and the real costs of the “Log Loss” cost function! During convergence the relative importance of the regularization term may therefore become bigger until the ratio to the standard costs reaches an eventual constant level. But how dominant will the regularization term get in the end?

Let us do some experiments with the MNIST dataset again! We fix some common parameters and conditions for our test runs:
As we saw in the last article we should normalize the input data. So, all of our numerical experiments below (with the exception of the last one) are done with standardized input data (using Scikit-Learn’s StandardScaler). In addition initial weights are all set according to the sqrt(nodes)-rule for all layers in the interval [-0.5*sqrt(1/num_nodes), 0.5*sqrt(1/num_nodes)], with num_nodes meaning the number of nodes in a layer. Other parameters, which we keep constant, are:

Parameters: learn_rate = 0.001, decrease_rate = 0.00001, mom_rate = 0.00005, n_size_mini_batch = 500, n_epochs = 800.

I added some statements to the method for cost calculation in order to save the relative part of the regularization terms with respect to the total costs of each mini-batch in a Numpy array and plot the evolution in the end. The changes are so simple that I omit showing the modified code.

A first look at the evolution of the relative contribution of regularization to the total loss of a mini-batch

How does the outcome of gradient descent look for standardized input data and a Lambda2-value of 0.1?

Lambda2 = 0.1
Results: acc_train: 0.999 , acc_test: 0.9714, convergence after ca. 600 epochs

We see that the regularization term actually dominates the total loss of a mini-batch at convergence. At least with our present parameter setting. In comparisoin to the total loss of the full training set the contribution is of course much smaller and typically below 1%.

A small Lambda term

Let us reduce the regularization term via setting Lambda = 0.01. We expect its initial contribution to the costs of a batch to be smaller then, but this does NOT mean that the ratio to the standard costs of the batch automatically shrinks significantly, too:

Lambda2 = 0.01
Results: acc_train: 1.0 , acc_test: 0.9656, convergence after ca. 350 epochs

Note the absolute scale of the costs in the plots! We ended up at a much lower level of the total loss of a batch! But the relative dominance of regularization at the point of convergence actually increased! However, this did not help us with the accuracy of our MLP-algorithm on the test data set – although we perfectly fit the training set by a 100% accuracy.

In the end this is what regularization is all about. We do not want a total overfitting, a perfect adaption of the grid to the training set. It will not help in the sense of getting a better general accuracy on other input data. A Lambda2 of 0.01 is much too small in our case!

Slightly bigger regularization with Lambda2 = 0.2

So lets enlarge Lambda2 a bit:
Lambda2 = 0.2
Results: acc_train: 0.9946 , acc_test: 0.9728, convergence after ca. 700 epochs

We get an improved accuracy!

Two other cases with significantly bigger Lambda2

Lambda2 = 0.4
Results: acc_train: 0.9858 , acc_test: 0.9693, convergence after ca. 600 epochs

Lambda2 = 0.8
Results: acc_train: 0.9705 , acc_test: 0.9588, convergence after ca. 400 epochs

OK, but in both cases we see a significant and systematic trend towards reduced accuracy values on the test data set with growing Lambda2-values > 0.2 for our chosen mini-batch size (500 samples).

Conclusion

We learned a bit about the impact of regularization today. Whatever the exact Lambda2-value – in the end the contribution of a regularization term becomes a significant part of the total loss of a mini-batch when we approached the total cost minimum. However, the factor Lambda2 must be chosen with a reasonable size to get an impact of regularization on the final minimum position in the weight-space! But then it will help to improve accuracy on general input data in comparison to overfitted solutions!

But we also saw that there is some balance to take care of: For an optimum of generalization AND accuracy you should neither make Lambda2 too small nor too big. In our case Lambda2 = 0.2 seems to be a reasonable and good choice. Might be different with other datasets.

All in all studying the impact of a variation of achieved accuracy with the factor for a Ridge regularization term seems to be a good investment of time in ML projects. We shall come back to this point already in the next articles of this series.

In the next article

A simple Python program for an ANN to cover the MNIST dataset – XIV – cluster detection in feature space

we shall start to work on cluster detection in the feature space of the MNIST data before using gradient descent.

 

A simple Python program for an ANN to cover the MNIST dataset – X – mini-batch-shuffling and some more tests

I continue my series on a Python code for a simple multi-layer perceptron [MLP]. During the course of the previous articles we have built a Python class “ANN” with methods to import the MNIST data set and handle forward propagation as well as error backward propagation [EBP]. We also had a deeper look at the mathematics of gradient descent and EBP for MLP training.

A simple program for an ANN to cover the Mnist dataset – IX – First Tests
A simple program for an ANN to cover the Mnist dataset – VIII – coding Error Backward Propagation
A simple program for an ANN to cover the Mnist dataset – VII – EBP related topics and obstacles
A simple program for an ANN to cover the Mnist dataset – VI – the math behind the „error back-propagation“
A simple program for an ANN to cover the Mnist dataset – V – coding the loss function
A simple program for an ANN to cover the Mnist dataset – IV – the concept of a cost or loss function
A simple program for an ANN to cover the Mnist dataset – III – forward propagation
A simple program for an ANN to cover the Mnist dataset – II – initial random weight values
A simple program for an ANN to cover the Mnist dataset – I – a starting point

The code modifications in the last article enabled us to perform a first test on the MNIST dataset. This test gave us some confidence in our training algorithm: It seemed to converge and produce weights which did a relatively good job on analyzing the MNIST images.

We saw a slight tendency of overfitting. But an accuracy level of 96.5% on the test dataset showed that the MLP had indeed “learned” something during training. We needed around 1000 epochs to come to this point.

However, there are a lot of parameters controlling our grid structure and the learning behavior. Such parameters are often called “hyper-parameters“. To get a better understanding of our MLP we must start playing around with such parameters. In this article we shall concentrate on the parameter for (regression) regularization (called Lambda2 in the parameter interface of our class ANN) and then start varying the node numbers on the layers.

But before we start new test runs we add a statistical element to the training – namely the variation of the composition of our mini-batches (see the last article).

General hint: In all of the test runs below we used 4 CPU cores with libOpenBlas on a Linux system with an I7 6700K CPU.

Shuffling the contents of the mini-batches

Let us add some more parameters to the interface of class “ANN”:

shuffle_batches = True
print_period = 20

The first parameter
shall control whether we vary the composition of the mini-batches with each epoch. The second parameter controls for which period of the epochs we print out some intermediate data (costs, averaged error of last mini-batch).

    def __init__(self, 
                 my_data_set = "mnist", 
                 n_hidden_layers = 1, 
                 ay_nodes_layers = [0, 100, 0], # array which should have as much elements as n_hidden + 2
                 n_nodes_layer_out = 10,  # expected number of nodes in output layer 
                                                  
                 my_activation_function = "sigmoid", 
                 my_out_function        = "sigmoid",   
                 my_loss_function       = "LogLoss",   
                 
                 n_size_mini_batch = 50,  # number of data elements in a mini-batch 
                 
                 n_epochs      = 1,
                 n_max_batches = -1,  # number of mini-batches to use during epochs - > 0 only for testing 
                                      # a negative value uses all mini-batches 
                 
                 lambda2_reg = 0.1,     # factor for quadratic regularization term 
                 lambda1_reg = 0.0,     # factor for linear regularization term 
                 
                 vect_mode = 'cols', 
                 
                 learn_rate = 0.001,        # the learning rate (often called epsilon in textbooks) 
                 decrease_const = 0.00001,  # a factor for decreasing the learning rate with epochs
                 mom_rate   = 0.0005,       # a factor for momentum learning
                 
                 shuffle_batches = True,    # True: we mix the data for mini-batches in the X-train set at the start of each epoch
                 
                 print_period = 20,         # number of epochs for which to print the costs and the averaged error
                 
                 figs_x1=12.0, figs_x2=8.0, 
                 legend_loc='upper right',
                 
                 b_print_test_data = True
                 
                 ):
        '''
        Initialization of MyANN
        Input: 
            data_set: type of dataset; so far only the "mnist", "mnist_784" datsets are known 
                      We use this information to prepare the input data and learn about the feature dimension. 
                      This info is used in preparing the size of the input layer.     
            n_hidden_layers = number of hidden layers => between input layer 0 and output layer n 
            
            ay_nodes_layers = [0, 100, 0 ] : We set the number of nodes in input layer_0 and the output_layer to zero 
                              Will be set to real number afterwards by infos from the input dataset. 
                              All other numbers are used for the node numbers of the hidden layers.
            n_nodes_out_layer = expected number of nodes in the output layer (is checked); 
                                this number corresponds to the number of categories NC = number of labels to be distinguished
            
            my_activation_function : name of the activation function to use 
            my_out_function : name of the "activation" function of the last layer which produces the output values 
            my_loss_function : name of the "cost" or "loss" function used for optimization 
            
            n_size_mini_batch : Number of elements/samples in a mini-batch of training data 
                                The number of mini-batches will be calculated from this
            
            n_epochs : number of epochs to calculate during training
            n_max_batches : > 0: maximum of mini-batches to use during training 
                      
      < 0: use all mini-batches  
            
            lambda_reg2:    The factor for the quadartic regularization term 
            lambda_reg1:    The factor for the linear regularization term 
            
            vect_mode: Are 1-dim data arrays (vctors) ordered by columns or rows ?
            
            learn rate :     Learning rate - definies by how much we correct weights in the indicated direction of the gradient on the cost hyperplane.
            decrease_const:  Controls a systematic decrease of the learning rate with epoch number 
            mom_const:       Momentum rate. Controls a mixture of the last with the present weight corrections (momentum learning)
            
            shuffle_batches: True => vary composition of mini-batches with each epoch
            
            print_period:    number of periods between printing out some intermediate data 
                             on costs and the averaged error of the last mini-batch   
                       
            
            figs_x1=12.0, figs_x2=8.0 : Standard sizing of plots , 
            legend_loc='upper right': Position of legends in the plots
            
            b_print_test_data: Boolean variable to control the print out of some tests data 
             
         '''
        
        # Array (Python list) of known input data sets 
        self._input_data_sets = ["mnist", "mnist_784", "mnist_keras"]  
        self._my_data_set = my_data_set
        
        # X, y, X_train, y_train, X_test, y_test  
            # will be set by analyze_input_data 
            # X: Input array (2D) - at present status of MNIST image data, only.    
            # y: result (=classification data) [digits represent categories in the case of Mnist]
        self._X       = None 
        self._X_train = None 
        self._X_test  = None   
        self._y       = None 
        self._y_train = None 
        self._y_test  = None
        
        # relevant dimensions 
        # from input data information;  will be set in handle_input_data()
        self._dim_sets     = 0  
        self._dim_features = 0  
        self._n_labels     = 0   # number of unique labels - will be extracted from y-data 
        
        # Img sizes 
        self._dim_img      = 0 # should be sqrt(dim_features) - we assume square like images  
        self._img_h        = 0 
        self._img_w        = 0 
        
        # Layers
        # ------
        # number of hidden layers 
        self._n_hidden_layers = n_hidden_layers
        # Number of total layers 
        self._n_total_layers = 2 + self._n_hidden_layers  
        # Nodes for hidden layers 
        self._ay_nodes_layers = np.array(ay_nodes_layers)
        # Number of nodes in output layer - will be checked against information from target arrays
        self._n_nodes_layer_out = n_nodes_layer_out
        
        # Weights 
        # --------
        # empty List for all weight-matrices for all layer-connections
        # Numbering : 
        # w[0] contains the weight matrix which connects layer 0 (input layer ) to hidden layer 1 
        # w[1] contains the weight matrix which connects layer 1 (input layer ) to (hidden?) layer 2 
        self._li_w = []
        
        # Arrays for encoded output labels - will be set in _encode_all_mnist_labels()
        # -------------------------------
        self._ay_onehot = None
        self._ay_oneval = None
        
        # Known Randomizer methods ( 0: np.random.randint, 1: np.random.uniform )  
        # ------------------
        self.__ay_known_randomizers = [0, 1]

        # Types of activation functions and output functions 
        # ------------------
        self.__ay_activation_functions = ["sigmoid"] # later also relu 
        self.__ay_output_functions     = ["sigmoid"] # later also 
softmax 
        
        # Types of cost functions 
        # ------------------
        self.__ay_loss_functions = ["LogLoss", "MSE" ] # later also other types of cost/loss functions  


        # the following dictionaries will be used for indirect function calls 
        self.__d_activation_funcs = {
            'sigmoid': self._sigmoid, 
            'relu':    self._relu
            }
        self.__d_output_funcs = { 
            'sigmoid': self._sigmoid, 
            'softmax': self._softmax
            }  
        self.__d_loss_funcs = { 
            'LogLoss': self._loss_LogLoss, 
            'MSE': self._loss_MSE
            }  
        # Derivative functions 
        self.__d_D_activation_funcs = {
            'sigmoid': self._D_sigmoid, 
            'relu':    self._D_relu
            }
        self.__d_D_output_funcs = { 
            'sigmoid': self._D_sigmoid, 
            'softmax': self._D_softmax
            }  
        self.__d_D_loss_funcs = { 
            'LogLoss': self._D_loss_LogLoss, 
            'MSE': self._D_loss_MSE
            }  
        
        
        # The following variables will later be set by _check_and set_activation_and_out_functions()            
        
        self._my_act_func  = my_activation_function
        self._my_out_func  = my_out_function
        self._my_loss_func = my_loss_function
        self._act_func = None    
        self._out_func = None    
        self._loss_func = None    
        
        # number of data samples in a mini-batch 
        self._n_size_mini_batch = n_size_mini_batch
        self._n_mini_batches = None  # will be determined by _get_number_of_mini_batches()

        # maximum number of epochs - we set this number to an assumed maximum 
        # - as we shall build a backup and reload functionality for training, this should not be a major problem 
        self._n_epochs = n_epochs
        
        # maximum number of batches to handle ( if < 0 => all!) 
        self._n_max_batches = n_max_batches
        # actual number of batches 
        self._n_batches = None

        # regularization parameters
        self._lambda2_reg = lambda2_reg
        self._lambda1_reg = lambda1_reg
        
        # parameter for momentum learning 
        self._learn_rate = learn_rate
        self._decrease_const = decrease_const
        self._mom_rate   = mom_rate
        self._li_mom = [None] *  self._n_total_layers
        
        # shuffle data in X_train? 
        self._shuffle_batches = shuffle_batches
        
        # epoch period for printing 
        self._print_period = print_period
        
        # book-keeping for epochs and mini-batches 
        # -------------------------------
        # range for epochs - will be set by _prepare-epochs_and_batches() 
        self._rg_idx_epochs = None
        # range for mini-batches 
        self._rg_idx_batches = None
        # dimension of the numpy arrays for book-keeping - will be set in _prepare_epochs_and_batches() 
        self._shape_epochs_batches = None    # (n_epochs, n_batches, 1) 

        # list for error values at outermost layer for minibatches and epochs during training
        # we use a numpy array here because we can redimension it
        self._ay_theta = None
        # list for cost values of mini-batches during training 
        # The list will later be split into sections for epochs 
        self._ay_costs = None
        
        
        # Data elements for back propagation
        # ----------------------------------
        
        # 2-dim array of partial derivatives of the elements of an additive cost function 
        # The derivative is taken with respect to the output results a_j = ay_ANN_out[j]
        # The array dimensions account for nodes and sampls of a 
mini_batch. The array will be set in function 
        # self._initiate_bw_propagation()
        self._ay_delta_out_batch = None
        

        # parameter to allow printing of some test data 
        self._b_print_test_data = b_print_test_data

        # Plot handling 
        # --------------
        # Alternatives to resize plots 
        # 1: just resize figure  2: resize plus create subplots() [figure + axes] 
        self._plot_resize_alternative = 1 
        # Plot-sizing
        self._figs_x1 = figs_x1
        self._figs_x2 = figs_x2
        self._fig = None
        self._ax  = None 
        # alternative 2 does resizing and (!) subplots() 
        self.initiate_and_resize_plot(self._plot_resize_alternative)        
        
        
        # ***********
        # operations 
        # ***********
        
        # check and handle input data 
        self._handle_input_data()
        # set the ANN structure 
        self._set_ANN_structure()
        
        # Prepare epoch and batch-handling - sets ranges, limits num of mini-batches and initializes book-keeping arrays
        self._rg_idx_epochs, self._rg_idx_batches = self._prepare_epochs_and_batches()
        
        # perform training 
        start_c = time.perf_counter()
        self._fit(b_print=True, b_measure_batch_time=False)
        end_c = time.perf_counter()
        print('\n\n ------') 
        print('Total training Time_CPU: ', end_c - start_c) 
        print("\nStopping program regularily")

 
Both parameters affect our method “_fit()” in the following way :

    ''' -- Method to set the number of batches based on given batch size -- '''
    def _fit(self, b_print = False, b_measure_batch_time = False):
        '''
        Parameters: 
            b_print:                 Do we print intermediate results of the training at all? 
            b_print_period:          For which period of epochs do we print? 
            b_measure_batch_time:    Measure CPU-Time for a batch
        '''
        rg_idx_epochs  = self._rg_idx_epochs 
        rg_idx_batches = self._rg_idx_batches
        if (b_print):    
            print("\nnumber of epochs = " + str(len(rg_idx_epochs)))
            print("max number of batches = " + str(len(rg_idx_batches)))
       
        # loop over epochs
        for idxe in rg_idx_epochs:
            if (b_print and (idxe % self._print_period == 0) ):
                print("\n ---------")
                print("\nStarting epoch " + str(idxe+1))
                
            # sinmple adaption of the learning rate 
            self._learn_rate /= (1.0 + self._decrease_const * idxe)
            
            # shuffle indices for a variation of the mini-batches with each epoch
            if self._shuffle_batches:
                shuffled_index = np.random.permutation(self._dim_sets)
                self._X_train, self._y_train, self._ay_onehot = self._X_train[shuffled_index], self._y_train[shuffled_index], self._ay_onehot[:, shuffled_index]
                
            
            # loop over mini-batches
            for idxb in rg_idx_batches:
                if b_measure_batch_time: 
                    start_0 = time.perf_counter()
                # deal with a mini-batch
                self._handle_mini_batch(num_batch = idxb, num_epoch=idxe, b_print_y_vals = False, b_print = False)
                if b_measure_batch_time: 
                    end_0 = time.perf_counter()
                    print('Time_CPU for batch ' + str(idxb+1), end_0 - start_0) 
                
            if (b_print and (idxe % self._print_period == 0) ):
                print("\ntotal costs of mini_batch = ", self._ay_costs[idxe, idxb])
      
          print("avg total error of mini_batch = ", self._ay_theta[idxe, idxb])
                
        return None

 

Results for shuffling the contents of the mini-batches

With shuffling we expect a slightly broader variation of the costs and the averaged error. But the accuracy should no change too much in the end. We start a new test run with the following parameters:

     ay_nodes_layers = [0, 70, 30, 0], 
     n_nodes_layer_out = 10,  
     my_loss_function = "LogLoss",
     n_size_mini_batch = 500,
     n_epochs = 1500, 
     n_max_batches = 2000,  # small values only for test runs
     lambda2_reg = 0.1, 
     lambda1_reg = 0.0,      
     vect_mode = 'cols', 
     learn_rate = 0.0001,
     decrease_const = 0.000001,
     mom_rate   = 0.00005,  
     shuffle_batches = True,
     print_period = 20,         
...

If we look at the intermediate printout for the last mini-batch of some epochs and compare it to the results given in the last article, we see a stronger variation in the costs and averaged error. The reason is that the composition of last mini-batch of an epoch changes with every epoch.

number of epochs = 1500
max number of batches = 120
---------
Starting epoch 1
total costs of mini_batch =  1757.7650929607967
avg total error of mini_batch =  0.17086198431410954
---------
Starting epoch 61
total costs of mini_batch =  511.7001121819204
avg total error of mini_batch =  0.030287362041332373
---------
Starting epoch 121
total costs of mini_batch =  435.2513093033654
avg total error of mini_batch =  0.023445601362614754
----------
Starting epoch 181
total costs of mini_batch =  361.8665831722295
avg total error of mini_batch =  0.018540003201911136
---------
Starting epoch 241
total costs of mini_batch =  293.31230634431023
avg total error of mini_batch =  0.0138237366634751
---------
Starting epoch 301
total costs of mini_batch =  332.70394217467936
avg total error of mini_batch =  0.017697548541363246
---------
Starting epoch 361
total costs of mini_batch =  249.26400606039937
avg total error of mini_batch =  0.011765164578232358
---------
Starting epoch 421
total costs of mini_batch =  240.0503762160913
avg total error of mini_batch =  0.011650843329895542
---------
Starting epoch 481
total costs of mini_batch =  222.89422430417295
avg total error of mini_batch =  0.011503859412784031
---------
Starting epoch 541
total costs of mini_batch =  200.1195962051405
avg total error of mini_batch =  0.009962020519104173
---------
tarting epoch 601
total costs of mini_batch =  206.74753168607685
avg total error of mini_batch =  0.01067995191155135
---------
Starting epoch 661
total costs of mini_batch =  171.14090717705736
avg total error of mini_batch =  0.0077091934178393105
---------
Starting epoch 721
total costs of mini_batch =  158.44967190977957
avg total error of mini_batch =  0.0070760922760890735
---------
Starting epoch 781
total costs of mini_batch =  165.4047453537401
avg total error of mini_batch =  0.008622788115637027
---------
Starting epoch 841
total costs of mini_batch =  140.52762105883642
avg total error of mini_batch =  0.0067360505574077766
---------
Starting epoch 901
total costs of mini_batch =  163.9117184790982
avg total error of mini_batch =  0.007431666926365192
---------
Starting epoch 961
total costs of mini_batch =  126.05539161877512
avg total error of mini_batch =  0.005982378079899406
---------
Starting epoch 1021
total costs of mini_batch =  114.89943308334199
avg total error of mini_batch =  0.005122976288751798
---------
Starting epoch 1081
total costs of mini_batch =  117.22051220670932
avg total error of mini_batch 
=  0.005185936692097749
---------
Starting epoch 1141
total costs of mini_batch =  140.88969853048422
avg total error of mini_batch =  0.007665464508660714
---------
Starting epoch 1201
total costs of mini_batch =  113.27223303239667
avg total error of mini_batch =  0.0059791015452599705
---------
Starting epoch 1261
total costs of mini_batch =  105.55343407063131
avg total error of mini_batch =  0.005000503315756879
---------
Starting epoch 1321
total costs of mini_batch =  130.48116668827038
avg total error of mini_batch =  0.006287118265324945
---------
Starting epoch 1381
total costs of mini_batch =  109.04042315247389
avg total error of mini_batch =  0.005874339148860562
---------
Starting epoch 1441
total costs of mini_batch =  121.01379412127089
avg total error of mini_batch =  0.0065105907117289944
---------
Starting epoch 1461
total costs of mini_batch =  103.08774822996196
avg total error of mini_batch =  0.005299079778792264
---------
Starting epoch 1481
total costs of mini_batch =  106.21334182056928
avg total error of mini_batch =  0.005343967730134955
-------
Total training Time_CPU:  1963.8792177759988

 
Note that the averaged error values result from averaging of the absolute values of the errors of all records in a batch! The small numbers are not due to a cancelling of positive by negative deviations. A contribution to the error at an output node is given by the absloute value of the difference between the predicted real output value and the encoded target output value. We then first calculate an average over all output nodes (=10) per record and then average these values over all records of a batch. Such an “averaged error” gives us a first indication of the accuracy level reached.

Note that this averaged error is not becoming a constant. The last values in the above list indicate that we do not get much better with the error than 0.0055 on the training data. Our approached minimum points on the various cost hyperplanes of the mini-batches obviously hop around the global minimum on the hyperplane of the total costs. One of the reasons is the varying composition of the mini-batches; another reason is that the cost hyperplanes of the various mini-batches themselves are different from the hyperplane of the total costs of all records of the test data set. We see the effects of a mixture of “batch gradient descent” and “stochastic gradient descent” here; i.e., we do not get rid of stochastic elements even if we are close to a global minimum.

Still we observe an overall convergent behavior at around 1050 epochs. There our curves get relatively flat.

Accuracy values are:

total accuracy for training data = 0.9914
total accuracy for test data        = 0.9611

So, this is pretty much the same as in our original run in the last article without data shuffling.

Dropping regularization

In the above run we had used a quadratic from of the regularization (often called Ridge regularization). In the next test run we shall drop regularization completely (Lambda2 = 0, Lambda1 = 0) and find out whether this hampers the generalization of our MLP and the resulting accuracy with respect to the test data set.

Resulting data for the last epochs of the test run are

Starting epoch 1001
total costs of mini_batch =  67.98542512352101
avg total error of mini_batch =  0.007449654093429594
---------
nStarting epoch 1051
total costs of mini_batch =  56.69195783294443
avg total error of mini_batch =  0.0063384571747725415
---------
Starting epoch 1101
total costs of mini_batch =  51.81035466850738
avg total error of mini_batch =  0.005939699354987233
---------
Starting epoch 1151
total costs of mini_batch =  52.23157716632318
avg total error of mini_batch =  0.006373981433882217
---------
Starting epoch 1201
total costs of mini_batch =  48.40298652277855
avg total error of mini_batch =  0.005653856253701317
---------
Starting epoch 1251
total costs of mini_batch =  45.00623540189525
avg total error of mini_batch =  0.005245339176038497
---------
Starting epoch 1301
total costs of mini_batch =  36.88409532579881
avg total error of mini_batch =  0.004600719544961844
---------
Starting epoch 1351
total costs of mini_batch =  36.53543045554845
avg total error of mini_batch =  0.003993852242709943
---------
Starting epoch 1401
total costs of mini_batch =  38.80422469954769
avg total error of mini_batch =  0.00464620714991714
---------
Starting epoch 1451
total costs of mini_batch =  42.39371261881638
avg total error of mini_batch =  0.005294796697150631
------
Total training Time_CPU:  2118.4527089519997

 
Note, by the way, that the absolute values of the costs depend on the regularization parameter; therefore we see somewhat lower values in the end than before. But the absolute cost values are not so important regarding the general convergence and the accuracy of the network reached.

We omit the plots and just give the accuracy values:

total accuracy for training data = 0.9874
total accuracy for test data        = 0.9514

We get a slight drop in accuracy for the test data set – small (1%), but notable. It is interesting the even the accuracy on the training data became influenced.

Why might it be interesting to check the impact of the regularization?

We learn from literature that regularization helps with overfitting. Another aspect discussed e.g. by Jörg Frochte in his book “Maschinelles Lernen” is, whether we have enough training data to determine the vast amount of weights in complicated networks. He suggests on page 190 of his book to consider the number of weights in an MLP and compare it with the number of available data points.

He suggests that one may run into trouble if the difference between the number of weights (number of degrees of freedom) and the number of data records (number of independent information data) becomes too big. However, his test example was a rather limited one and for regression not classification. He also notes that if the data are well distributed may not be as big as assumed. If one thinks about it, one may also come to the question whether the real amount of data provided by the records is not by a factor of 10 larger – as we use 10 output values per record ….

Anyway, I think it is worthwhile to have a look at regularization.

Enlarging the regularization factor

We double the value of Lambda2: Lambda2 = 0.2.

Starting epoch 1251
total costs of mini_batch =  128.00827405482318
avg total error of mini_batch =  0.007276206815511017
---------
Starting epoch 1301
total costs of mini_batch =  107.62983581797556
avg total error of mini_batch =  0.005535653858885446
---------
Starting epoch 1351
total costs of mini_batch =  107.83630092292944
avg total error of mini_batch =  0.
005446805325519184
---------
Starting epoch 1401
total costs of mini_batch =  119.7648277329852
avg total error of mini_batch =  0.00729466852297802
---------
Starting epoch 1451
total costs of mini_batch =  106.74254206278933
avg total error of mini_batch =  0.005343124456075227

 
We get a slight improvement of the accuracy compared to our first run with data shuffling:

total accuracy for training data = 0.9950
total accuracy for test data        = 0.964

So, regularization does have its advantages. I recommend to investigate the impact of this parameter closely, if you need to get the “last percentages” in generalization and accuracy for a given MLP-model.

Enlarging node numbers

We have 60000 data records in the training set. In our example case we needed to fix around 784*70 + 70*30 + 30*10 = 57280 weight values. This is pretty close to the total amount of training data (60000). What happens if we extend the number of weights beyond the number of training records?
E.g. 784*100 + 100*50 + 50*10 = 83900. Do we get some trouble?

The results are:

          
Starting epoch 1151
total costs of mini_batch =  109.77341617599176
avg total error of mini_batch =  0.005494982077591186
---------
Starting epoch 1201
total costs of mini_batch =  113.5293680548904
avg total error of mini_batch =  0.005352117137100675
---------
Starting epoch 1251
total costs of mini_batch =  116.26371170820423
avg total error of mini_batch =  0.0072335516486698
---------
Starting epoch 1301
total costs of mini_batch =  99.7268420386945
avg total error of mini_batch =  0.004850817052601995
---------
Starting epoch 1351
total costs of mini_batch =  101.16579732551999
avg total error of mini_batch =  0.004831600835072556
---------
Starting epoch 1401
total costs of mini_batch =  98.45208584213253
avg total error of mini_batch =  0.004796133492821962
---------
Starting epoch 1451
total costs of mini_batch =  99.279344780807
avg total error of mini_batch =  0.005289728162205425
------
Total training Time_CPU:  2159.5880855739997

 

Ooops, there appears a glitch in the data around epoch 1250. Such things happen! So, we should have a look at the graphs before we decide to take the weights of a special epoch for our MLP model!

But in the end, i.e. with the weights at epoch 1500 the accuracy values are:

total accuracy for training data = 0.9962
total accuracy for test data        = 0.9632

So, we were NOT punished by extending our network, but we gained nothing worth the effort.

Now, let us go up with node numbers much more: 300 and 100 => 784*300 + 300*100 + 100*10 = 266200; ie. substantially more individual weights than training samples! First with Lambda2 = 0.2:

Starting epoch 1201
total costs of mini_batch =  104.4420759423322
avg total error of mini_batch =  0.0037985801450468246
---------
Starting epoch 1251
total costs of mini_batch =  102.80878647657674
avg total error of mini_batch =  0.003926855904089417
---------
Starting epoch 1301
total costs of mini_batch =  100.01189950545002
avg total error of mini_batch =  0.0037743225368465773
---------
Starting epoch 1351
total 
costs of mini_batch =  97.34294880936079
avg total error of mini_batch =  0.0035513092392408865
---------
Starting epoch 1401
total costs of mini_batch =  93.15432903284587
avg total error of mini_batch =  0.0032916082743134206
---------
Starting epoch 1451
total costs of mini_batch =  89.79127326241868
avg total error of mini_batch =  0.0033628384147655283
------
Total training Time_CPU:  4254.479082876998

 

total accuracy for training data = 0.9987
total accuracy for test data        = 0.9630

So , much CPU-time for no gain!

Now, what happens if we set Lambda2 = 0? We get:

total accuracy for training data = 0.9955
total accuracy for test data        = 0.9491

This is a small change around 1.4%! I would say:

In the special case of MNIST, a MLP-network with 2 hidden layers and a small learning rate we see neither a major problem regarding the amount of available data, nor a dramatic impact of the regularization. Regularization brings around a 1% gain in accuracy.

Reducing the node number on the hidden layers

Another experiment could be to reduce the number of nodes on the hidden layers. Let us go down to 40 nodes on L1 and 20 on L2, then to 30 and 20. Lambda2 is set to 0.2 in both runs. acc1 in the following listing means the accuracy for the training data, acc2 for the test data.

The results are:

40 nodes on L1, 20 nodes on L2, 1500 epochs => 1600 sec, acc1 = 0.9898, acc2 = 0.9578
30 nodes on L1, 20 nodes on L2, 1800 epochs => 1864 sec, acc1 = 0.9861, acc2 = 0.9513

We loose around 1% in accuracy!

The plot for 30, 20 nodes on layers L1, L2 shows that we got a convergence only beyond 1500 epochs.

Working with just one hidden layer

To get some more indications regarding efficiency let us now turn to networks with just one layer.
We investigate three situations with the following node numbers on the hidden layer: 100, 50, 30

The plot for 100 nodes on the hidden layer; we get convergence at around 1050 epochs.

Interestingly the CPU time for 100 nodes is with 1850 secs not smaller than for the network with 70 and 30 nodes on the hidden layers. As the dominant matrices are the ones connecting layer L0 and layer L1 this is quite understandable. (Note the CPU time also depends on the consumption of other jobs on the system.

The plots for 50 and 30 nodes on the hidden layer; we get convergence at around 1450 epochs. The
CPU time for 1500 epochs goes down to 1500 sec and XXX sec, respectively.

Plot for 50 nodes on the hidden layer:

Plot for 30 nodes:

We get the following accuracy values:

100 nodes, 1950 sec (1500 epochs), acc1 = 0.9947, acc2 = 0.9619,
50 nodes, 1600 sec (1500 epochs), acc1 = 0.9880, acc2 = 0.9566,
30 nodes, 1450 sec (1500 epochs), acc1 = 0.9780, acc2 = 0.9436

Well, now we see a drop in the accuracy by around 2% compared to our best cases. You have to decide yourself whether the gain in CPU time is worth it.

Note, by the way, that the accuracy value for 50 nodes is pretty close to the value S. Rashka got in his book “Python Machine Learning”. If you compare such values with your own runs be aware of the rather small learning rate (0.0001) and momentum rates (0.00005) I used. You can probably become faster with smaller learning rates. But then you may need another type of adaption for the learning rate compared to our simple linear one.

Conclusion

We saw that our original guess of a network with 2 hidden layers with 70 and 30 nodes was not a bad one. A network with just one layer with just 50 nodes or below does not give us the same accuracy. However, we neither saw a major improvement if we went to grids with 300 nodes on layer L1 and 100 nodes on layer L2. Despite some discrepancy between the number of weights in comparison to the number of test records we saw no significant loss in accuracy either – with or without regularization.
We also learned that we should use regularization (here of the quadratic type) to get the last 1% to 2% bit of accuracy on the test data in our special MNIST case.

In the next article

A simple program for an ANN to cover the Mnist dataset – XI – confusion matrix

we shall have a closer look at those MNIST number images where our MLP got problems.

 

A simple Python program for an ANN to cover the MNIST dataset – VIII – coding Error Backward Propagation

I continue with my series on a Python program for coding small “Multilayer Perceptrons” [MLPs].

A simple program for an ANN to cover the Mnist dataset – VII – EBP related topics and obstacles
A simple program for an ANN to cover the Mnist dataset – VI – the math behind the „error back-propagation“
A simple program for an ANN to cover the Mnist dataset – V – coding the loss function
A simple program for an ANN to cover the Mnist dataset – IV – the concept of a cost or loss function
A simple program for an ANN to cover the Mnist dataset – III – forward propagation
A simple program for an ANN to cover the Mnist dataset – II – initial random weight values
A simple program for an ANN to cover the Mnist dataset – I – a starting point

After all the theoretical considerations of the last two articles we now start coding again. Our objective is to extend our methods for training the MLP on the MNIST dataset by methods which perform the “error back propagation” and the correction of the weights. The mathematical prescriptions were derived in the following PDF:
My PDF on “The math behind EBP”

When you study the new code fragments below remember a few things:
We are prepared to use mini-batches. Therefore, the cost functions will be calculated over the data records of each batch and all the matrix operations for back propagation will cover all batch-records in parallel. Training means to loop over epochs and mini-batches – in pseudo-code:

  • Loop over epochs
    1. adjust learning rate,
    2. check for convergence criteria,
    3. Shuffle all data records in the test data set and build new mini-batches
  • Loop over mini-batches
    1. Perform forward propagation for all records of the mini-batch
    2. calculate and save the total cost value for each mini-batch
    3. calculate and save an averaged error on the output layer for each mini-batch
    4. perform error backward propagation on all records of the mini-batch to get the gradient of the cost function with respect to all weights
    5. adjust all weights on all layers

As discussed in the last article: The cost hyperplane changes a bit with each mini-batch. If there is a good mixture of records in a batch then the form of its specific cost hyperplane will (hopefully) resemble the form of an overall cost hyperplane, but it will not be the same. By the second step in the outer loop we want to avoid that the same data records always get an influence on the gradients at the same position in the correction procedure. Both statistical elements help a bit to overcome dominant records and a non-equal distribution of test records. If we had only pictures for number 3 at the end of our MNIST data set we
may start learning “3” representations very well, but not other numbers. Statistical variation also helps to avoid side minima on the overall cost hyperplane for all data records of the test set.

We shall implement the second step and third step in the epoch loop in the next article – when we are sure that the training algorithm works as expected. So, at the moment we will stop our training only after a given number of epochs.

More input parameters

In the first articles we had build an __init__()-method to parameterize a training run. We have to include three more parameters to control the backward propagation.

learn_rate = 0.001, # the learning rate (often called epsilon in textbooks)
decrease_const = 0.00001, # a factor for decreasing the learning rate with epochs
mom_rate = 0.0005, # a factor for momentum learning

The first parameter controls by how much we change weights with the help of gradient values. See formula (93) in the PDF of article VI (you find the Link to the latest version in the last section of this article). The second parameter will give us an option to decrease the learning rate with the number of training epochs. Note that a constant decrease rate only makes sense, if we can be relatively sure that we do not end up in a side minimum of the cost function.

The third parameter is interesting: It will allow us to mix the presently calculated weight correction with the correction term from the last step. So to say: We extend the “momentum” of the last correction into the next correction. This helps us not to follow indicated direction changes on the cost hyperplanes too fast.

Some hygienic measures regarding variables

In the already written parts of the code we have used a prefix “ay_” for all variables which represent some vector or array like structure – including Python lists and Numpy arrays. For back propagation coding it will be more important to distinguish between lists and arrays. So, I changed the variable prefix for some important Python lists from “ay_” to “li_”. (I shall do it for all lists used in a further version). In addition I have changed the prefix for Python ranges to “rg_”. These changes will affect the contents and the interface of some methods. You will notice when we come to these methods.

The changed __input__()-method

Our modified __init__() function now looks like this:

    def __init__(self, 
                 my_data_set = "mnist", 
                 n_hidden_layers = 1, 
                 ay_nodes_layers = [0, 100, 0], # array which should have as much elements as n_hidden + 2
                 n_nodes_layer_out = 10,  # expected number of nodes in output layer 
                                                  
                 my_activation_function = "sigmoid", 
                 my_out_function        = "sigmoid",   
                 my_loss_function       = "LogLoss",   
                 
                 n_size_mini_batch = 50,  # number of data elements in a mini-batch 
                 
                 n_epochs      = 1,
                 n_max_batches = -1,  # number of mini-batches to use during epochs - > 0 only for testing 
                                      # a negative value uses all mini-batches 
                 
                 lambda2_reg = 0.1,     # factor for quadratic regularization term 
                 lambda1_reg = 0.0,     # factor for linear regularization term 
                 
                 vect_mode = 'cols', 
                 
                 learn_rate = 0.001,        # the learning rate (often called epsilon in textbooks) 
                 decrease_const = 0.00001,  # a factor for decreasing the learning rate with epochs
                 mom_rate  
 = 0.0005,       # a factor for momentum learning
                 
                 figs_x1=12.0, figs_x2=8.0, 
                 legend_loc='upper right',
                 
                 b_print_test_data = True
                 
                 ):
        '''
        Initialization of MyANN
        Input: 
            data_set: type of dataset; so far only the "mnist", "mnist_784" datsets are known 
                      We use this information to prepare the input data and learn about the feature dimension. 
                      This info is used in preparing the size of the input layer.     
            n_hidden_layers = number of hidden layers => between input layer 0 and output layer n 
            
            ay_nodes_layers = [0, 100, 0 ] : We set the number of nodes in input layer_0 and the output_layer to zero 
                              Will be set to real number afterwards by infos from the input dataset. 
                              All other numbers are used for the node numbers of the hidden layers.
            n_nodes_out_layer = expected number of nodes in the output layer (is checked); 
                                this number corresponds to the number of categories NC = number of labels to be distinguished
            
            my_activation_function : name of the activation function to use 
            my_out_function : name of the "activation" function of the last layer which produces the output values 
            my_loss_function : name of the "cost" or "loss" function used for optimization 
            
            n_size_mini_batch : Number of elements/samples in a mini-batch of training data 
                                The number of mini-batches will be calculated from this
            
            n_epochs : number of epochs to calculate during training
            n_max_batches : > 0: maximum of mini-batches to use during training 
                            < 0: use all mini-batches  
            
            lambda_reg2:    The factor for the quadartic regularization term 
            lambda_reg1:    The factor for the linear regularization term 
            
            vect_mode: Are 1-dim data arrays (vctors) ordered by columns or rows ?
            
            learn rate :     Learning rate - definies by how much we correct weights in the indicated direction of the gradient on the cost hyperplane.
            decrease_const:  Controls a systematic decrease of the learning rate with epoch number 
            mom_const:       Momentum rate. Controls a mixture of the last with the present weight corrections (momentum learning)
            
            figs_x1=12.0, figs_x2=8.0 : Standard sizing of plots , 
            legend_loc='upper right': Position of legends in the plots
            
            b_print_test_data: Boolean variable to control the print out of some tests data 
             
         '''
        
        # Array (Python list) of known input data sets 
        self._input_data_sets = ["mnist", "mnist_784", "mnist_keras"]  
        self._my_data_set = my_data_set
        
        # X, y, X_train, y_train, X_test, y_test  
            # will be set by analyze_input_data 
            # X: Input array (2D) - at present status of MNIST image data, only.    
            # y: result (=classification data) [digits represent categories in the case of Mnist]
        self._X       = None 
        self._X_train = None 
        self._X_test  = None   
        self._y       = None 
        self._y_train = None 
        self._y_test  = None
        
        # relevant dimensions 
        # from input data information;  will be set in handle_input_data()
        self._dim_sets     = 0  
        self._dim_features = 0  
        self._n_labels     = 0   # number of unique labels - will be extracted from y-data 
        
      
  # Img sizes 
        self._dim_img      = 0 # should be sqrt(dim_features) - we assume square like images  
        self._img_h        = 0 
        self._img_w        = 0 
        
        # Layers
        # ------
        # number of hidden layers 
        self._n_hidden_layers = n_hidden_layers
        # Number of total layers 
        self._n_total_layers = 2 + self._n_hidden_layers  
        # Nodes for hidden layers 
        self._ay_nodes_layers = np.array(ay_nodes_layers)
        # Number of nodes in output layer - will be checked against information from target arrays
        self._n_nodes_layer_out = n_nodes_layer_out
        
        # Weights 
        # --------
        # empty List for all weight-matrices for all layer-connections
        # Numbering : 
        # w[0] contains the weight matrix which connects layer 0 (input layer ) to hidden layer 1 
        # w[1] contains the weight matrix which connects layer 1 (input layer ) to (hidden?) layer 2 
        self._li_w = []
        
        # Arrays for encoded output labels - will be set in _encode_all_mnist_labels()
        # -------------------------------
        self._ay_onehot = None
        self._ay_oneval = None
        
        # Known Randomizer methods ( 0: np.random.randint, 1: np.random.uniform )  
        # ------------------
        self.__ay_known_randomizers = [0, 1]

        # Types of activation functions and output functions 
        # ------------------
        self.__ay_activation_functions = ["sigmoid"] # later also relu 
        self.__ay_output_functions     = ["sigmoid"] # later also softmax 
        
        # Types of cost functions 
        # ------------------
        self.__ay_loss_functions = ["LogLoss", "MSE" ] # later also othr types of cost/loss functions  


        # the following dictionaries will be used for indirect function calls 
        self.__d_activation_funcs = {
            'sigmoid': self._sigmoid, 
            'relu':    self._relu
            }
        self.__d_output_funcs = { 
            'sigmoid': self._sigmoid, 
            'softmax': self._softmax
            }  
        self.__d_loss_funcs = { 
            'LogLoss': self._loss_LogLoss, 
            'MSE': self._loss_MSE
            }  
        # Derivative functions 
        self.__d_D_activation_funcs = {
            'sigmoid': self._D_sigmoid, 
            'relu':    self._D_relu
            }
        self.__d_D_output_funcs = { 
            'sigmoid': self._D_sigmoid, 
            'softmax': self._D_softmax
            }  
        self.__d_D_loss_funcs = { 
            'LogLoss': self._D_loss_LogLoss, 
            'MSE': self._D_loss_MSE
            }  
        
        
        # The following variables will later be set by _check_and set_activation_and_out_functions()            
        
        self._my_act_func  = my_activation_function
        self._my_out_func  = my_out_function
        self._my_loss_func = my_loss_function
        self._act_func = None    
        self._out_func = None    
        self._loss_func = None    
        
        # number of data samples in a mini-batch 
        self._n_size_mini_batch = n_size_mini_batch
        self._n_mini_batches = None  # will be determined by _get_number_of_mini_batches()

        # maximum number of epochs - we set this number to an assumed maximum 
        # - as we shall build a backup and reload functionality for training, this should not be a major problem 
        self._n_epochs = n_epochs
        
        # maximum number of batches to handle ( if < 0 => all!) 
        self._n_max_batches = n_max_batches
        # actual number of batches 
        self._n_batches = None

        # regularization parameters
        self._lambda2_reg = lambda2_reg
        self._
lambda1_reg = lambda1_reg
        
        # parameter for momentum learning 
        self._learn_rate = learn_rate
        self._decrease_const = decrease_const
        self._mom_rate   = mom_rate
        self._li_mom = [None] *  self._n_total_layers
        
        # book-keeping for epochs and mini-batches 
        # -------------------------------
        # range for epochs - will be set by _prepare-epochs_and_batches() 
        self._rg_idx_epochs = None
        # range for mini-batches 
        self._rg_idx_batches = None
        # dimension of the numpy arrays for book-keeping - will be set in _prepare_epochs_and_batches() 
        self._shape_epochs_batches = None    # (n_epochs, n_batches, 1) 

        # list for error values at outermost layer for minibatches and epochs during training
        # we use a numpy array here because we can redimension it
        self._ay_theta = None
        # list for cost values of mini-batches during training 
        # The list will later be split into sections for epochs 
        self._ay_costs = None
        
        
        # Data elements for back propagation
        # ----------------------------------
        
        # 2-dim array of partial derivatives of the elements of an additive cost function 
        # The derivative is taken with respect to the output results a_j = ay_ANN_out[j]
        # The array dimensions account for nodes and sampls of a mini_batch. The array will be set in function 
        # self._initiate_bw_propagation()
        self._ay_delta_out_batch = None
        

        # parameter to allow printing of some test data 
        self._b_print_test_data = b_print_test_data

        # Plot handling 
        # --------------
        # Alternatives to resize plots 
        # 1: just resize figure  2: resize plus create subplots() [figure + axes] 
        self._plot_resize_alternative = 1 
        # Plot-sizing
        self._figs_x1 = figs_x1
        self._figs_x2 = figs_x2
        self._fig = None
        self._ax  = None 
        # alternative 2 does resizing and (!) subplots() 
        self.initiate_and_resize_plot(self._plot_resize_alternative)        
        
        
        # ***********
        # operations 
        # ***********
        
        # check and handle input data 
        self._handle_input_data()
        # set the ANN structure 
        self._set_ANN_structure()
        
        # Prepare epoch and batch-handling - sets ranges, limits num of mini-batches and initializes book-keeping arrays
        self._rg_idx_epochs, self._rg_idx_batches = self._prepare_epochs_and_batches()
        
        # perform training 
        start_c = time.perf_counter()
        self._fit(b_print=True, b_measure_batch_time=False)
        end_c = time.perf_counter()
        print('\n\n ------') 
        print('Total training Time_CPU: ', end_c - start_c) 
        print("\nStopping program regularily")
        sys.exit()

The extended method _set_ANN_structure()

I do not change method “_handle_input_data()”. However, I extend method “def _set_ANN_structure()” by a statement to initialize a list with momentum matrices for all layers.

    '''-- Method to set ANN structure --''' 
    def _set_ANN_structure(self):
        # check consistency of the node-number list with the number of hidden layers (n_hidden)
        self._check_layer_and_node_numbers()
        # set node numbers for the input layer and the output layer
        self._set_nodes_for_input_output_layers() 
        self._show_node_numbers() 
        
        # create the weight matrix between input and first hidden layer 
        self._create_WM_Input() 
        # create weight matrices between the 
hidden layers and between tha last hidden and the output layer 
        self._create_WM_Hidden() 
        
        # initialize momentum differences
        self._create_momentum_matrices()
        #print("\nLength of li_mom = ", str(len(self._li_mom)))
        
        # check and set activation functions 
        self._check_and_set_activation_and_out_functions()
        self._check_and_set_loss_function()
        
        return None

 
The following box shows the changed functions _create_WM_Input(), _create_WM_Hidden() and the new function _create_momentum_matrices():

    '''-- Method to create the weight matrix between L0/L1 --'''
    def _create_WM_Input(self):
        '''
        Method to create the input layer 
        The dimension will be taken from the structure of the input data 
        We need to fill self._w[0] with a matrix for conections of all nodes in L0 with all nodes in L1
        We fill the matrix with random numbers between [-1, 1] 
        '''
        # the num_nodes of layer 0 should already include the bias node 
        num_nodes_layer_0 = self._ay_nodes_layers[0]
        num_nodes_with_bias_layer_0 = num_nodes_layer_0 + 1 
        num_nodes_layer_1 = self._ay_nodes_layers[1] 
        
        # fill the matrix with random values 
        #rand_low  = -1.0
        #rand_high = 1.0
        rand_low  = -0.5
        rand_high = 0.5
        rand_size = num_nodes_layer_1 * (num_nodes_with_bias_layer_0) 
        
        randomizer = 1 # method np.random.uniform   
        
        w0 = self._create_vector_with_random_values(rand_low, rand_high, rand_size, randomizer)
        w0 = w0.reshape(num_nodes_layer_1, num_nodes_with_bias_layer_0)
        
        # put the weight matrix into array of matrices 
        self._li_w.append(w0)
        print("\nShape of weight matrix between layers 0 and 1 " + str(self._li_w[0].shape))
        
#
    '''-- Method to create the weight-matrices for hidden layers--''' 
    def _create_WM_Hidden(self):
        '''
        Method to create the weights of the hidden layers, i.e. between [L1, L2] and so on ... [L_n, L_out] 
        We fill the matrix with random numbers between [-1, 1] 
        '''
        
        # The "+1" is required due to range properties ! 
        rg_hidden_layers = range(1, self._n_hidden_layers + 1, 1)

        # for random operation 
        rand_low  = -1.0
        rand_high = 1.0
        
        for i in rg_hidden_layers: 
            print ("Creating weight matrix for layer " + str(i) + " to layer " + str(i+1) )
            
            num_nodes_layer = self._ay_nodes_layers[i] 
            num_nodes_with_bias_layer = num_nodes_layer + 1 
            
            # the number of the next layer is taken without the bias node!
            num_nodes_layer_next = self._ay_nodes_layers[i+1]
            
            # assign random values  
            rand_size = num_nodes_layer_next * num_nodes_with_bias_layer   
            
            randomizer = 1 # np.random.uniform
            
            w_i_next = self._create_vector_with_random_values(rand_low, rand_high, rand_size, randomizer)   
            w_i_next = w_i_next.reshape(num_nodes_layer_next, num_nodes_with_bias_layer)
            
            # put the weight matrix into our array of matrices 
            self._li_w.append(w_i_next)
            print("Shape of weight matrix between layers " + str(i) + " and " + str(i+1) + " = " + str(self._li_w[i].shape))
#
    '''-- Method to create and initialize matrices for momentum learning (differences) '''
    def _create_momentum_matrices(self):
        rg_layers = range(0, self._n_total_layers - 1)
        for i in rg_layers: 
r
            self._li_mom[i] = np.zeros(self._li_w[i].shape)
            #print("shape of li_mom[" + str(i) + "] = ", self._li_mom[i].shape)

 

The modified functions _fit() and _handle_mini_batch()

The _fit()-function is modified to include a systematic decrease of the learning rate.

    ''' -- Method to set the number of batches based on given batch size -- '''
    def _fit(self, b_print = False, b_measure_batch_time = False):
        
        rg_idx_epochs  = self._rg_idx_epochs 
        rg_idx_batches = self._rg_idx_batches
        if (b_print):    
            print("\nnumber of epochs = " + str(len(rg_idx_epochs)))
            print("max number of batches = " + str(len(rg_idx_batches)))
       
        # loop over epochs
        for idxe in rg_idx_epochs:
            if (b_print):
                print("\n ---------")
                print("\nStarting epoch " + str(idxe+1))
                
                self._learn_rate /= (1.0 + self._decrease_const * idxe)
            
            # loop over mini-batches
            for idxb in rg_idx_batches:
                if (b_print):
                    print("\n ---------")
                    print("\nDealing with mini-batch " + str(idxb+1))
                if b_measure_batch_time: 
                    start_0 = time.perf_counter()
                # deal with a mini-batch
                self._handle_mini_batch(num_batch = idxb, num_epoch=idxe, b_print_y_vals = False, b_print = False)
                if b_measure_batch_time: 
                    end_0 = time.perf_counter()
                    print('Time_CPU for batch ' + str(idxb+1), end_0 - start_0) 
                
                #if idxb == 100: 
                #    sys.exit() 
        
        return None

 
Note that the number of epochs is determined by an external parameter as an upper limit of the range “rg_idx_epochs”.

Method “_handle_mini_batch()” requires several changes: First we define lists which are required to save matrix data of the backward propagation. And, of course, we call a method to perform the BW propagation (see step 6 in the code). Some statements print shapes, if required. At step 7 of the code we correct the weights by using the learning rate and the calculated gradient of the loss function.

Note, that we mix the correction evaluated at the last batch-record with the correction evaluated for the present record! This corresponds to a simple form of momentum learning. We then have to save the present correction values, of course. Note that the list for momentum correction “li_mom” is, therefore, not deleted at the end of a mini-batch treatment !

In addition to saving the total costs per mini-batch we now also save a mean error at the output level. The average is done by the help of Numpy’s function numpy.average() for matrices. Remember, we build the average over errors at all output nodes and all records of the mini-batch.


    ''' -- Method to deal with a batch -- '''
    def _handle_mini_batch(self, num_batch = 0, num_epoch = 0, b_print_y_vals = False, b_print = False, b_keep_bw_matrices = True):
        '''
        For each batch we keep the input data array Z and the output data A (output of activation function!) 
        for all layers in Python lists
        We can use this as input variables in function calls - mutable variables are handled by reference values !
        We receive the A and Z data from propagation functions and proceed them to cost and gradient calculation functions
        
        As an initial step we define the Python lists li_Z_
in_layer and li_A_out_layer 
        and fill in the first input elements for layer L0  
        
        Forward propagation:
        --------------------
        li_Z_in_layer : List of layer-related 2-dim matrices for input values z at each node (rows) and all batch-samples (cols).
        li_A_out_layer: List of layer-related 2-dim matrices for output alues z at each node (rows) and all batch-samples (cols).
                        The output is created by Phi(z), where Phi represents an activation or output function 
        
        Note that the matrices in ay_A_out will be permanently extended by a row (over all samples) 
        to account for a bias node of each inner layer. This happens during FW propagation. 
        
        Note that the matrices ay_Z_in will be temporarily extended by a row (over all samples) 
        to account for a bias node of each inner layer. This happens during BW propagation.
        
        Backward propagation:
        -------------------- 
        li_delta_out:  Startup matrix for _out_delta-values at the outermost layer 
        li_grad_layer: List of layer-related matrices with gradient values for the correction of the weights 
        
        Depending on parameter "b_keep_bw_matrices" we keep 
            - a list of layer-related matrices D with values for the derivatives of the act./output functions
            - a list of layer-related matrices for the back propagated delta-values 
        in lists during back propagation. This can support error analysis. 
        
        All matrices in the lists are 2 dimensional with dimensions for nodes (rows) and training samples (cols) 
        All these lists be deleted at the end of the function to accelerate garbadge handling
        
        Input parameters: 
        ----------------
        num_epoch:     Number of present epoch
        num_batch:    Number of present mini-batch 
        '''
        # Layer-related lists to be filled with 2-dim Numpy matrices during FW propagation
        # ********************************************************************************
        li_Z_in_layer  = [None] * self._n_total_layers # List of matrices with z-input values for each layer; filled during FW-propagation
        li_A_out_layer = li_Z_in_layer.copy()          # List of matrices with results of activation/output-functions for each layer; filled during FW-propagation
        li_delta_out   = li_Z_in_layer.copy()          # Matrix with out_delta-values at the outermost layer 
        li_delta_layer = li_Z_in_layer.copy()          # List of the matrices for the BW propagated delta values 
        li_D_layer     = li_Z_in_layer.copy()          # List of the derivative matrices D containing partial derivatives of the activation/ouput functions 
        li_grad_layer  = li_Z_in_layer.copy()          # List of the matrices with gradient values for weight corrections
        
        if b_print: 
            len_lists = len(li_A_out_layer)
            print("\nnum_epoch = ", num_epoch, "  num_batch = ", num_batch )
            print("\nhandle_mini_batch(): length of lists = ", len_lists)
            self._info_point_print("handle_mini_batch: point 1")
        
        # Print some infos
        # ****************
        if b_print:
            self._print_batch_infos()
            self._info_point_print("handle_mini_batch: point 2")
        
        # Major steps for the mini-batch during one epoch iteration 
        # **********************************************************
        
        # Step 0: List of indices for data records in the present mini-batch
        # ******
        ay_idx_batch = self._ay_mini_batches[num_batch]
        
        # Step 1: Special preparation of the Z-input to the MLP's input Layer L0
        # ******
        # Layer L0: Fill in the input vector for the ANN's input layer L0 
        li_
Z_in_layer[0] = self._X_train[ay_idx_batch] # numpy arrays can be indexed by an array of integers
        if b_print:
            print("\nPropagation : Shape of X_in = li_Z_in_layer = ", li_Z_in_layer[0].shape)           
            #print("\nidx, expected y_value of Layer L0-input :")           
            #for idx in self._ay_mini_batches[num_batch]:
            #    print(str(idx) + ', ' + str(self._y_train[idx]) )
            self._info_point_print("handle_mini_batch: point 3")
        
        # Step 2: Layer L0: We need to transpose the data of the input layer 
        # *******
        ay_Z_in_0T       = li_Z_in_layer[0].T
        li_Z_in_layer[0] = ay_Z_in_0T
        if b_print:
            print("\nPropagation : Shape of transposed X_in = li_Z_in_layer = ", li_Z_in_layer[0].shape)           
            self._info_point_print("handle_mini_batch: point 4")
        
        # Step 3: Call forward propagation method for the present mini-batch of training records
        # *******
        # this function will fill the ay_Z_in- and ay_A_out-lists with matrices per layer
        self._fw_propagation(li_Z_in = li_Z_in_layer, li_A_out = li_A_out_layer, b_print = b_print) 
        
        if b_print:
            ilayer = range(0, self._n_total_layers)
            print("\n ---- ")
            print("\nAfter propagation through all " + str(self._n_total_layers) + " layers: ")
            for il in ilayer:
                print("Shape of Z_in of layer L" + str(il) + " = " + str(li_Z_in_layer[il].shape))
                print("Shape of A_out of layer L" + str(il) + " = " + str(li_A_out_layer[il].shape))
                if il < self._n_total_layers-1:
                    print("Shape of W of layer L" + str(il) + " = " + str(self._li_w[il].shape))
                    print("Shape of Mom of layer L" + str(il) + " = " + str(self._li_mom[il].shape))
            self._info_point_print("handle_mini_batch: point 5")
        
        
        # Step 4: Cost calculation for the mini-batch 
        # ********
        ay_y_enc = self._ay_onehot[:, ay_idx_batch]
        ay_ANN_out = li_A_out_layer[self._n_total_layers-1]
        # print("Shape of ay_ANN_out = " + str(ay_ANN_out.shape))
        
        total_costs_batch = self._calculate_loss_for_batch(ay_y_enc, ay_ANN_out, b_print = False)
        # we add the present cost value to the numpy array 
        self._ay_costs[num_epoch, num_batch] = total_costs_batch
        if b_print:
            print("\n total costs of mini_batch = ", self._ay_costs[num_epoch, num_batch])
            self._info_point_print("handle_mini_batch: point 6")
        print("\n total costs of mini_batch = ", self._ay_costs[num_epoch, num_batch])
        
        # Step 5: Avg-error for later plotting 
        # ********
        # mean "error" values - averaged over all nodes at outermost layer and all data sets of a mini-batch 
        ay_theta_out = ay_y_enc - ay_ANN_out
        if (b_print): 
            print("Shape of ay_theta_out = " + str(ay_theta_out.shape))
        ay_theta_avg = np.average(np.abs(ay_theta_out)) 
        self._ay_theta[num_epoch, num_batch] = ay_theta_avg 
        
        if b_print:
            print("\navg total error of mini_batch = ", self._ay_theta[num_epoch, num_batch])
            self._info_point_print("handle_mini_batch: point 7")
        print("avg total error of mini_batch = ", self._ay_theta[num_epoch, num_batch])
        
        
        # Step 6: Perform gradient calculation via back propagation of errors
        # ******* 
        self._bw_propagation( ay_y_enc = ay_y_enc, 
                              li_Z_in = li_Z_in_layer, 
                              li_A_out = li_A_out_layer, 
                              li_delta_out = li_delta_out, 
                              li_delta = li_delta_
layer,
                              li_D = li_D_layer, 
                              li_grad = li_grad_layer, 
                              b_print = b_print,
                              b_internal_timing = False 
                              ) 
        
        
        # Step 7: Adjustment of weights  
        # *******
        rg_layer=range(0, self._n_total_layers -1)
        for N in rg_layer:
            delta_w_N = self._learn_rate * li_grad_layer[N]
            self._li_w[N] -= ( delta_w_N + (self._mom_rate * self._li_mom[N]) )
            # save momentum
            self._li_mom[N] = delta_w_N
        
        # try to accelerate garbage handling
        # **************
        if len(li_Z_in_layer) > 0:
            del li_Z_in_layer
        if len(li_A_out_layer) > 0:
            del li_A_out_layer
        if len(li_delta_out) > 0:
            del li_delta_out
        if len(li_delta_layer) > 0:
            del li_delta_layer
        if len(li_D_layer) > 0:
            del li_D_layer
        if len(li_grad_layer) > 0:
            del li_grad_layer
            
        return None

 

Forward Propagation

The method for forward propagation remains unchanged in its structure. We only changed the prefix for the Python lists.

    ''' -- Method to handle FW propagation for a mini-batch --'''
    def _fw_propagation(self, li_Z_in, li_A_out, b_print= False):
        
        b_internal_timing = False
        
        # index range of layers 
        #    Note that we count from 0 (0=>L0) to E L(=>E) / 
        #    Careful: during BW-propgation we may need a correct indexing of lists filled during FW-propagation
        ilayer = range(0, self._n_total_layers-1)

        # propagation loop
        # ***************
        for il in ilayer:
            if b_internal_timing: start_0 = time.perf_counter()
            
            if b_print: 
                print("\nStarting propagation between L" + str(il) + " and L" + str(il+1))
                print("Shape of Z_in of layer L" + str(il) + " (without bias) = " + str(li_Z_in[il].shape))
            
            # Step 1: Take input of last layer and apply activation function 
            # ******
            if il == 0: 
                A_out_il = li_Z_in[il] # L0: activation function is identity 
            else: 
                A_out_il = self._act_func( li_Z_in[il] ) # use real activation function 
            
            # Step 2: Add bias node
            # ****** 
            A_out_il = self._add_bias_neuron_to_layer(A_out_il, 'row')
            # save in array     
            li_A_out[il] = A_out_il
            if b_print: 
                print("Shape of A_out of layer L" + str(il) + " (with bias) = " + str(li_A_out[il].shape))
            
            # Step 3: Propagate by matrix operation
            # ****** 
            Z_in_ilp1 = np.dot(self._li_w[il], A_out_il) 
            li_Z_in[il+1] = Z_in_ilp1
            
            if b_internal_timing: 
                end_0 = time.perf_counter()
                print('Time_CPU for layer propagation L' + str(il) + ' to L' + str(il+1), end_0 - start_0) 
        
        # treatment of the last layer 
        # ***************************
        il = il + 1
        if b_print:
            print("\nShape of Z_in of layer L" + str(il) + " = " + str(li_Z_in[il].shape))
        A_out_il = self._out_func( li_Z_in[il] ) # use the output function 
        li_A_out[il] = A_out_il
        if b_print:
            print("Shape of A_out of last layer L" + str(il) + " = " + str(li_A_out[il].shape))
        
        return None

 
nAddendum, 15.05.2020:
We shall later learn that the treatment of bias neurons can be done more efficiently. The present way of coding it reduces performance – especially at the input layer. See the article series starting with
MLP, Numpy, TF2 – performance issues – Step I – float32, reduction of back propagation
for more information. At the present stage of our discussion we are, however, more interested in getting a working code first – and not so much in performance optimization.

Methods for Error Backward Propagation

In contrast to the recipe given in my PDF on the EBP-math we cannot calculate the matrices with the derivatives of the activation functions “ay_D” in advance for all layers. The reason was discussed in the last article VII: Some matrices have to be intermediately adjusted for a bias-neuron, which is ignored in the analysis of the PDF.

The resulting code of our method for EBP looks like given below:

 
    ''' -- Method to handle error BW propagation for a mini-batch --'''
    def _bw_propagation(self, 
                        ay_y_enc, li_Z_in, li_A_out, 
                        li_delta_out, li_delta, li_D, li_grad, 
                        b_print = True, b_internal_timing = False):
        
        # List initialization: All parameter lists or arrays are filled or to be filled by layers 
        # Note: the lists li_Z_in, li_A_out were already filled by _fw_propagation() for the present batch 
        
        # Initiate BW propagation - provide delta-matrices for outermost layer
        # *********************** 
        # Input Z at outermost layer E  (4 layers -> layer 3)
        ay_Z_E = li_Z_in[self._n_total_layers-1]
        # Output A at outermost layer E (was calculated by output function)
        ay_A_E = li_A_out[self._n_total_layers-1]
        
        # Calculate D-matrix (derivative of output function) at outmost the layer - presently only D_sigmoid 
        ay_D_E = self._calculate_D_E(ay_Z_E=ay_Z_E, b_print=b_print )
        
        # Get the 2 delta matrices for the outermost layer (only layer E has 2 delta-matrices)
        ay_delta_E, ay_delta_out_E = self._calculate_delta_E(ay_y_enc=ay_y_enc, ay_A_E=ay_A_E, ay_D_E=ay_D_E, b_print=b_print) 
        
        # We check the shapes
        shape_theory = (self._n_nodes_layer_out, self._n_size_mini_batch)
        if (b_print and ay_delta_E.shape != shape_theory):
            print("\nError: Shape of ay_delta_E is wrong:")
            print("Shape = ", ay_delta_E.shape, "  ::  should be = ", shape_theory )
        if (b_print and ay_D_E.shape != shape_theory):
            print("\nError: Shape of ay_D_E is wrong:")
            print("Shape = ", ay_D_E.shape, "  ::  should be = ", shape_theory )
        
        # add the matrices to their lists ; li_delta_out gets only one element 
        idxE = self._n_total_layers - 1
        li_delta_out[idxE] = ay_delta_out_E # this happens only once
        li_delta[idxE]     = ay_delta_E
        li_D[idxE]         = ay_D_E
        li_grad[idxE]      = None    # On the outermost layer there is no gradient ! 
        
        if b_print:
            print("bw: Shape delta_E = ", li_delta[idxE].shape)
            print("bw: Shape D_E = ", ay_D_E.shape)
            self._info_point_print("bw_propagation: point bw_1")
        
        
        # Loop over all layers in reverse direction 
        # ******************************************
        # index range of target layers N in BW direction (starting with E-1 => 4 layers -> layer 2))
        if b_print:
            range_N_bw_layer_test = reversed(range(0,
 self._n_total_layers-1))   # must be -1 as the last element is not taken 
            rg_list = list(range_N_bw_layer_test) # Note this exhausts the range-object
            print("range_N_bw_layer = ", rg_list)
        
        range_N_bw_layer = reversed(range(0, self._n_total_layers-1))   # must be -1 as the last element is not taken 
        
        # loop over layers 
        for N in range_N_bw_layer:
            if b_print:
                print("\n N (layer) = " + str(N) +"\n")
            # start timer 
            if b_internal_timing: start_0 = time.perf_counter()
            
            # Back Propagation operations between layers N+1 and N 
            # *******************************************************
            # this method handles the special treatment of bias nodes in Z_in, too
            ay_delta_N, ay_D_N, ay_grad_N = self._bw_prop_Np1_to_N( N=N, li_Z_in=li_Z_in, li_A_out=li_A_out, li_delta=li_delta, b_print=False )
            
            if b_internal_timing: 
                end_0 = time.perf_counter()
                print('Time_CPU for BW layer operations ', end_0 - start_0) 
            
            # add matrices to their lists 
            li_delta[N] = ay_delta_N
            li_D[N]     = ay_D_N
            li_grad[N]= ay_grad_N
            #sys.exit()
        
        return

 

We first handle the necessary matrix evaluations for the outermost layer. We use two helper functions there to calculate the derivative of the output function with respect to the a-term [ _calculate_D_E() ] and to calculate the values for the “delta“-terms at all nodes and for all records [ _calculate_delta_E() ] according to the prescription in the PDF:

    
    ''' -- Method to calculate the matrix with the derivative values of the output function at outermost layer '''
    def _calculate_D_E(self, ay_Z_E, b_print= True):
        '''
        This method calculates and returns the D-matrix for the outermost layer
        The D matrix contains derivatives of the output function with respect to local input "z_j" at outermost nodes. 
        
        Returns
        ------
        ay_D_E:    Matrix with derivative values of the output function 
                   with respect to local z_j valus at the nodes of the outermost layer E
        Note: This is a 2-dim matrix over layer nodes and training samples of the mini-batch
        '''
        if self._my_out_func == 'sigmoid':
            ay_D_E = self._D_sigmoid(ay_Z=ay_Z_E)
        
        else:
            print("The derivative for output function " + self._my_out_func + " is not known yet!" )
            sys.exit()
        
        return ay_D_E

    ''' -- Method to calculate the delta_E matrix as a starting point of the backward propagation '''
    def _calculate_delta_E(self, ay_y_enc, ay_A_E, ay_D_E, b_print= False):
        '''
        This method calculates and returns the 2 delta-matrices for the outermost layer 
        
        Returns
        ------
        delta_E:     delta_matrix of the outermost layer (indicated by E)
        delta_out:   delta_out matrix => elements are local derivative values of the cost function 
                     with respect to the output "a_j" at an outermost node  
                     !!! delta_out will only be returned if calculable !!!
        
        Note: these are 2-dim matrices over layer nodes and training samples of the mini-batch
        '''
        
        if self._my_loss_func == 'LogLoss':
            # Calculate delta_S_E directly to avoid problems with zero denominators
            ay_delta_E = ay_A_E - ay_y_enc
            # delta_out is fetched but may be None 
            ay_delta_out, ay_D_
numerator, ay_D_denominator = self._D_loss_LogLoss(ay_y_enc, ay_A_E, b_print = False)
            
            # To be done: Analyze critical values in D_denominator 
            
            # Release variables explicitly 
            del ay_D_numerator
            del ay_D_denominator
            
        
        if self._my_loss_func == 'MSE':
            # First calculate delta_out and then the delta_E
            delta_out = self._D_loss_MSE(ay_y_enc, ay_A_E, b_print=False)
            # calculate delta_E via matrix multiplication 
            ay_delta_E = delta_out * ay_D_E
                    
        return ay_delta_E, ay_delta_out

 

Further required helper methods to calculate the cost functions and related derivatives are :

    ''' method to calculate the logistic regression loss function '''
    def _loss_LogLoss(self, ay_y_enc, ay_ANN_out, b_print = False):
        '''
        Method which calculates LogReg loss function in a vectorized form on multidimensional Numpy arrays 
        '''
        b_test = False

        if b_print:
            print("From LogLoss: shape of ay_y_enc =  " + str(ay_y_enc.shape))
            print("From LogLoss: shape of ay_ANN_out =  " + str(ay_ANN_out.shape))
            print("LogLoss: ay_y_enc = ", ay_y_enc) 
            print("LogLoss: ANN_out = \n", ay_ANN_out) 
            print("LogLoss: log(ay_ANN_out) =  \n", np.log(ay_ANN_out) )

        # The following means an element-wise (!) operation between matrices of the same shape!
        Log1 = -ay_y_enc * (np.log(ay_ANN_out))
        # The following means an element-wise (!) operation between matrices of the same shape!
        Log2 = (1 - ay_y_enc) * np.log(1 - ay_ANN_out)
        
        # the next operation calculates the sum over all matrix elements 
        # - thus getting the total costs for all mini-batch elements 
        cost = np.sum(Log1 - Log2)
        
        #if b_print and b_test:
            # Log1_x = -ay_y_enc.dot((np.log(ay_ANN_out)).T)
            # print("From LogLoss: L1 =   " + str(L1))
            # print("From LogLoss: L1X =  " + str(L1X))
        
        if b_print: 
            print("From LogLoss: cost =  " + str(cost))
        
        # The total costs is just a number (scalar)
        return cost
#
    ''' method to calculate the derivative of the logistic regression loss function 
        with respect to the output values '''
    def _D_loss_LogLoss(self, ay_y_enc, ay_ANN_out, b_print = False):
        '''
        This function returns the out_delta_S-matrix which is required to initialize the 
        BW propagation (EBP) 
        Note ANN_out is the A_out-list element ( a 2-dim matrix) for the outermost layer 
        In this case we have to take care of denominators = 0 
        '''
        D_numerator = ay_ANN_out - ay_y_enc
        D_denominator = -(ay_ANN_out - 1.0) * ay_ANN_out
        n_critical = np.count_nonzero(D_denominator < 1.0e-8)
        if n_critical > 0:
            delta_s_out = None
        else:
            delta_s_out = np.divide(D_numerator, D_denominator)
        return delta_s_out, D_numerator, D_denominator
#
    ''' method to calculate the MSE loss function '''
    def _loss_MSE(self, ay_y_enc, ay_ANN_out, b_print = False):
        '''
        Method which calculates LogReg loss function in a vectorized form on multidimensional Numpy arrays 
        '''
        if b_print:
            print("From loss_MSE: shape of ay_y_enc =  " + str(ay_y_enc.shape))
            print("From loss_MSE: shape of ay_ANN_out =  " + str(ay_ANN_out.shape))
            #print("LogReg: ay_y_enc = ", ay_y_enc) 
            #print("LogReg: ANN_out = \n", ay_
ANN_out) 
            #print("LogReg: log(ay_ANN_out) =  \n", np.log(ay_ANN_out) )
        
        cost = 0.5 * np.sum( np.square( ay_y_enc - ay_ANN_out ) )

        if b_print: 
            print("From loss_MSE: cost =  " + str(cost))
        
        return cost
#
    ''' method to calculate the derivative of the MSE loss function 
        with respect to the output values '''
    def _D_loss_MSE(self, ay_y_enc, ay_ANN_out, b_print = False):
        '''
        This function returns the out_delta_S - matrix which is required to initialize the 
        BW propagation (EBP) 
        Note ANN_out is the A_out-list element ( a 2-dim matrix) for the outermost layer
        In this case the output is harmless (no critical denominator) 
        '''
        delta_s_out = ay_ANN_out - ay_y_enc
        return delta_s_out

 
You see that we are a bit careful to avoid zero denominators for the Logarithmic loss function in all of our helper functions.

The check statements for shapes can be eliminated in a future version when we are sure that everything works correctly. Keeping the layer specific matrices during the handling of a mini-batch will be also good for potentially required error analysis in the beginning. In the end we only may keep the gradient-matrices and the layer specific matrices required to process the local calculations during back propagation.

Then we turn to loop over all other layers down to layer L0. The matrix operation to be done for all these layers are handled in a further method:

    
    ''' -- Method to calculate the BW-propagated delta-matrix and the gradient matrix to/for layer N '''
    def _bw_prop_Np1_to_N(self, N, li_Z_in, li_A_out, li_delta, b_print=False):
        '''
        BW-error-propagation bewtween layer N+1 and N 
        Inputs: 
            li_Z_in:  List of input Z-matrices on all layers - values were calculated during FW-propagation
            li_A_out: List of output A-matrices - values were calculated during FW-propagation
            li_delta: List of delta-matrices - values for outermost ölayer E to layer N+1 should exist 
        
        Returns: 
            ay_delta_N - delta-matrix of layer N (required in subsequent steps)
            ay_D_N     - derivative matrix for the activation function on layer N 
            ay_grad_N  - matrix with gradient elements of the cost fnction with respect to the weights on layer N 
        '''
        
        if b_print:
            print("Inside _bw_prop_Np1_to_N: N = " + str(N) )
        
        # Prepare required quantities - and add bias neuron to ay_Z_in 
        # ****************************
        
        # Weight matrix meddling betwen layer N and N+1 
        ay_W_N = self._li_w[N]
        shape_W_N   = ay_W_N.shape # due to bias node first dim is 1 bigger than Z-matrix 
        if b_print:
            print("shape of W_N = ", shape_W_N )

        # delta-matrix of layer N+1
        ay_delta_Np1 = li_delta[N+1]
        shape_delta_Np1 = ay_delta_Np1.shape 

        # !!! Add intermediate row (for bias) to Z_N !!!
        ay_Z_N = li_Z_in[N]
        shape_Z_N_orig = ay_Z_N.shape
        ay_Z_N = self._add_bias_neuron_to_layer(ay_Z_N, 'row')
        shape_Z_N = ay_Z_N.shape # dimensions should fit now with W- and A-matrix 
        
        # Derivative matrix for the activation function (with extra bias node row)
        #    can only be calculated now as we need the z-values
        ay_D_N = self._calculate_D_N(ay_Z_N)
        shape_D_N = ay_D_N.shape 
        
        ay_A_N = li_A_out[N]
        shape_A_N = ay_A_N.shape
        
        # print shapes 
        if b_print:
            print("shape of W_N = ", shape_W_N)
            print("
shape of delta_(N+1) = ", shape_delta_Np1)
            print("shape of Z_N_orig = ", shape_Z_N_orig)
            print("shape of Z_N = ", shape_Z_N)
            print("shape of D_N = ", shape_D_N)
            print("shape of A_N = ", shape_A_N)
        
        
        # Propagate delta
        # **************
        if li_delta[N+1] is None:
            print("BW-Prop-error:\n No delta-matrix found for layer " + str(N+1) ) 
            sys.exit()
            
        # Check shapes for np.dot()-operation - here for element [0] of both shapes - as we operate with W.T !
        if ( shape_W_N[0] != shape_delta_Np1[0]): 
            print("BW-Prop-error:\n shape of W_N [", shape_W_N, "]) does not fit shape of delta_N+1 [", shape_delta_Np1, "]" )
            sys.exit() 
        
        # intermediate delta 
        # ~~~~~~~~~~~~~~~~~~
        ay_delta_w_N = ay_W_N.T.dot(ay_delta_Np1)
        shape_delta_w_N = ay_delta_w_N.shape
        
        # Check shapes for element wise *-operation !
        if ( shape_delta_w_N != shape_D_N ): 
            print("BW-Prop-error:\n shape of delta_w_N [", shape_delta_w_N, "]) does not fit shape of D_N [", shape_D_N, "]" )
            sys.exit() 
        
        # final delta 
        # ~~~~~~~~~~~
        ay_delta_N = ay_delta_w_N * ay_D_N
        # reduce dimension again 
        ay_delta_N = ay_delta_N[1:, :]
        shape_delta_N = ay_delta_N.shape
        
        # Check dimensions again - ay_delta_N.shape should fit shape_Z_in_orig
        if shape_delta_N != shape_Z_N_orig: 
            print("BW-Prop-error:\n shape of delta_N [", shape_delta_N, "]) does not fit original shape Z_in_N [", shape_Z_N_orig, "]" )
            sys.exit() 
        
        if N > 0:
            shape_W_Nm1 = self._li_w[N-1].shape 
            if shape_delta_N[0] != shape_W_Nm1[0] : 
                print("BW-Prop-error:\n shape of delta_N [", shape_delta_N, "]) does not fit shape of W_Nm1 [", shape_W_Nm1, "]" )
                sysexit() 
        
        
        # Calculate gradient
        # ********************
        #     required for all layers down to 0 
        # check shapes 
        if shape_delta_Np1[1] != shape_A_N[1]:
            print("BW-Prop-error:\n shape of delta_Np1 [", shape_delta_Np1, "]) does not fit shape of A_N [", shape_A_N, "] for matrix multiplication" )
            sys.exit() 
        
        # calculate gradient             
        ay_grad_N = np.dot(ay_delta_Np1, ay_A_N.T)
        
        # regularize gradient (!!!! without adding bias nodes in the L1, L2 sums) 
        ay_grad_N[:, 1:] += (self._li_w[N][:, 1:] * self._lambda2_reg + np.sign(self._li_w[N][:, 1:]) * self._lambda1_reg) 
        
        #
        # Check shape 
        shape_grad_N = ay_grad_N.shape
        if shape_grad_N != shape_W_N:
            print("BW-Prop-error:\n shape of grad_N [", shape_grad_N, "]) does not fit shape of W_N [", shape_W_N, "]" )
            sys.exit() 
        
        # print shapes 
        if b_print:
            print("shape of delta_N = ", shape_delta_N)
            print("shape of grad_N = ", shape_grad_N)
            
            print(ay_grad_N)
        
        return ay_delta_N, ay_D_N, ay_grad_N

 
This function does more or less exactly what we have requested by our theoretical analysis in the last two articles. Note the intermediate handling of bias nodes! Note also that bias nodes are NOT regarded in regularization terms L1 and L2! The function to calculate the derivative of the activation function is:

   
#
    ''' -- Method to calculate the matrix with the derivative values of the output function at outermost layer '''

    def _calculate_D_N(self, ay_Z_N, b_print= False):
        '''
        This method calculates and returns the D-matrix for the outermost layer
        The D matrix contains derivatives of the output function with respect to local input "z_j" at outermost nodes. 
        
        Returns
        ------
        ay_D_E:    Matrix with derivative values of the output function 
                   with respect to local z_j valus at the nodes of the outermost layer E
        Note: This is a 2-dim matrix over layer nodes and training samples of the mini-batch
        '''
        if self._my_out_func == 'sigmoid':
            ay_D_E = self._D_sigmoid(ay_Z = ay_Z_N)
        
        else:
            print("The derivative for output function " + self._my_out_func + " is not known yet!" )
            sys.exit()
        
        return ay_D_E
 

 

The methods to calculate regularization terms for the loss function are:

   
#
    ''' method do calculate the quadratic regularization term for the loss function '''
    def _regularize_by_L2(self, b_print=False): 
        '''
        The L2 regularization term sums up all quadratic weights (without the weight for the bias) 
        over the input and all hidden layers (but not the output layer)
        The weight for the bias is in the first column (index 0) of the weight matrix - 
        as the bias node's output is in the first row of the output vector of the layer 
        '''
        ilayer = range(0, self._n_total_layers-1) # this excludes the last layer 
        L2 = 0.0
        for idx in ilayer:
            L2 += (np.sum( np.square(self._li_w[idx][:, 1:])) ) 
        L2 *= 0.5 * self._lambda2_reg
        if b_print: 
            print("\nL2: total L2 = " + str(L2) )
        return L2 
#
    ''' method do calculate the linear regularization term for the loss function '''
    def _regularize_by_L1(self, b_print=False): 
        '''
        The L1 regularization term sums up all weights (without the weight for the bias) 
        over the input and all hidden layers (but not the output layer
        The weight for the bias is in the first column (index 0) of the weight matrix - 
        as the bias node's output is in the first row of the output vector of the layer 
        '''
        ilayer = range(0, self._n_total_layers-1) # this excludes the last layer 
        L1 = 0.0
        for idx in ilayer:
            L1 += np.sum(np.abs( self._li_w[idx][:, 1:]))
        L1 *= 0.5 * self._lambda1_reg
        if b_print:
            print("\nL1: total L1 = " + str(L1))
        return L1 

 
Addendum, 15.05.2020:
Also the BW-propagation code presented here will later be the target of optimization steps. We shall see that it – despite working correctly – can be criticized regarding efficiency at several points. See again the article series starting with
MLP, Numpy, TF2 – performance issues – Step I – float32, reduction of back propagation.

Conclusion

We have extended our set of methods quite a bit. At the core of the operations we perform matrix operations which are supported by the Openblas library on a Linux system with multiple CPU cores. In the next article

A simple program for an ANN to cover the Mnist dataset – IX – First Tests

we shall test the convergence of our training for the MNIST data set. We shall see that a MLP with two hidden layers with 70 and 30 nodes can give us a convergence
of the averaged relative error down to 0.006 after 1000 epochs on the test data. However, we have to analyze such results for overfitting. Stay tuned …

Links

My PDF on “The math behind EBP”