# Using embeddings, collaborative filtering with Deep Learning to analyse T20 players

There is a school of thought which considers that total runs scored and strike rate for a batsman, or total wickets taken and economy rate for a bowler, do not tell the whole story. This is true to a fair extent. The runs scored or the wickets taken could have been against weaker teams and hence the runs, strike rate or the wickets and economy rate alone do not capture all the performance details of the batsman or bowler. A technique to determine the performance of batsmen against different bowlers and identify the batsman’s possible performance even against bowlers he/she has not yet faced could be done with collaborative filtering. Collaborative filtering, with embeddings can also be used to group players with similar characteristics. Similarly, we could also identify the performance of bowlers versus different batsmen. Hence we need to look at average runs, SR and total wickets, ER with the lens of batsmen, bowlers against similar opposition. This is where collaborative filtering is useful.

The table below shows the performance of all batsman against all bowlers in the table below. The row in the table below is the batsman and the column is the bowler, with the value in the cell is the total Runs scored by the batsman against the bowler in all matches. Note the values are 0 for batsmen who have not yet faced specific bowlers. The table is fairly sparse.

Table A

Similarly, we can compute the performance of all bowlers against all batsmen as in the table below. Here the row is the bowler, the column batsman and the value in the cell is the number of times the bowler got the batsman’s wicket. As before the data is sparsely populated

This problem of computing batsman’s performance against bowlers or vice versa, is identical to the user vs movie rating problem used in collaborative filtering. For e.g we could consider

This above problem depicted could be computed using collaborative filtering with embeddings. We could assign sequential numbers for the batsmen from 1 to M, and for the bowlers from 1 to N. The total runs scored could be represented only for the rows where there are values. One way to solve this problem in Machine Learning is to use One Hot Encoding (OHE), where we assign values for each row and each column and map the values of the table with values of the cell for each combination. But this would take a enormous computation time and memory. The solution to this is use vector embeddings. Here embeddings could be used for capturing the sparse tensors between the batsmen, bowlers, runs scored or vice versa between bowlers against batsmen and the wickets taken. We only need to consider the cells for which values exist. An embedding is a relatively low-dimensional space, into which you can translate high-dimensional vectors. An embedding captures some of the semantics of the input by placing semantically similar inputs close together in the embedding space.

a) To compute bowler performances and identify similarities between bowlers the following embedding in the Deep Learning Network was used

To compute batsmen similarities a similar Deep Learning network for bowler vs batsmen is used

I had earlier created another post Player Performance Estimation using AI Collaborative Filtering for batsman and bowler recommendation, using R package Recommender Lab. However, I was not too happy with the results I got with this R package. When I searched the net for material on using embeddings for collaborative filtering, most of material on the web on movie lens or word2vec are repetitive and have no new material. Finally, this short video lecture from Developer Google on Embeddings provided the most clarity.

I have created 4 Colab notebooks to identify player similarities (recommendations)

a) Batsman similarities IPL

b) Batsman similarities T20

c) Bowler similarities IPL

d) Bowler similarities T20

For creating the model I have used all the data for T20 and IPL from so that I get the best results. The data is from Cricsheet. I have also used Google’s Embeddings Projector to display batsman and bowler embedding to and to group similar players

All the Colab notebooks and the data associated with the code are available in Github. Feel free to download and execute them. See if you get better performance. I tried a wide variety of hyperparameters – learning rate, width and depth of nodes per layer, number of layers, gradient methods etc.

You can download all the code & data from Github at embeddings

A) Batsman Recommender IPL (BatsmanRecommenderIPLA.ipynb)

Steps for creating the model

a) Upload bowler vs batsmen with times wicket was taken for batsman. This will be a sparse matrix

d) Minimise loss for wickets taken for the bowler using SGD

e) Display embeddings of similar batsmen using Tensorboard projector

Upload data file
2. Remove rows where wickets = 0

import io
print(df2.shape)
df2 = df2.loc[df2['wicketTaken']!= 0]
print(df2.shape)

df6


Out[14]:

b) Create integer dictionaries for batsmen & bowlers

bowlers = df3["bowler1"].unique().tolist()
bowlers
# Create dictionary of bowler to index
bowlers2index = {x: i for i, x in enumerate(bowlers)}
bowlers2index
#Create dictionary of index tp bowler
index2bowlers = {i: x for i, x in enumerate(bowlers)}
index2bowlers

batsmen = df3["batsman1"].unique().tolist()
batsmen
# Create dictionary of batsman to index
batsmen2index = {x: i for i, x in enumerate(batsmen)}
batsmen2index
# Create dictionary of index to batsman
index2batsmen = {i: x for i, x in enumerate(batsmen)}
index2batsmen

#Map bowler, batsman to respective indices
df3["bowler"] = df3["bowler1"].map(bowlers2index)
df3["batsman"] = df3["batsman1"].map(batsmen2index)
df3
num_bowlers =len(bowlers2index)
num_batsmen = len(batsmen2index)
df3["wicketTaken"] = df3["wicketTaken"].values.astype(np.float32)
df3
# min and max ratings will be used to normalize the ratings later
min_wicketTaken = min(df3["wicketTaken"])
max_wicketTaken = max(df3["wicketTaken"])

print(
"Number of bowlers: {}, Number of batsmen: {}, Min wicketsTaken: {}, Max wicketsTaken: {}".format(
num_bowlers, num_batsmen, min_wicketTaken, max_wicketTaken
)
)

df3
df6
df31=pd.concat([df3,df6],axis=1)
df31

d) Create a Tensorflow/Keras deep learning mode. Minimise using Mean Squared Error using Stochastic Gradient Descent. I used ‘dropouts’ to regularise the model to keep validation loss within limits

tf.random.set_seed(4)
vector_size=len(batsmen2index)

df4=df31[['bowler','batsman','wicketTaken','balls','runsConceded','ER']]
df4
train_dataset = df4.sample(frac=0.9,random_state=0)
test_dataset = df4.drop(train_dataset.index)

train_dataset1 = train_dataset[['bowler','batsman','balls','runsConceded','ER']]
test_dataset1 = test_dataset[['bowler','batsman','balls','runsConceded','ER']]
train_stats = train_dataset1.describe()
train_stats = train_stats.transpose()
#print(train_stats)

train_labels = train_dataset.pop('wicketTaken')
test_labels = test_dataset.pop('wicketTaken')

# Create a Deep Learning model with keras
model = tf.keras.Sequential([
tf.keras.layers.Embedding(vector_size,16,input_length=5),
tf.keras.layers.Flatten(),
keras.layers.Dropout(.2),
keras.layers.Dense(16),

keras.layers.Dense(8,activation=tf.nn.relu),

keras.layers.Dense(4,activation=tf.nn.relu),
keras.layers.Dense(1)
])

# Print the model summary
#model.summary()
# Use the Adam optimizer with a learning rate of 0.01
#optimizer=keras.optimizers.RMSprop(learning_rate=0.01, rho=0.2, momentum=0.2, epsilon=1e-07)
#optimizer=keras.optimizers.SGD(learning_rate=.009,momentum=0.1) - Works without dropout
optimizer=keras.optimizers.SGD(learning_rate=.01,momentum=0.1)

model.compile(loss='mean_squared_error',
optimizer=optimizer,
)

# Setup the training parameters
#model.compile(loss='binary_crossentropy',optimizer='rmsprop',metrics=['accuracy'])
# Create a model
history=model.fit(
train_dataset1, train_labels,batch_size=32,
epochs=40, validation_data = (test_dataset1,test_labels), verbose=1)

e) Plot losses

f) Predict wickets that will be taken by bowlers against random batsmen


df5= df4[['bowler','batsman','balls','runsConceded','ER']]
test1 = df5.sample(n=10)
test1.shape
for i in range(test1.shape[0]):
print('Bowler :', index2bowlers.get(test1.iloc[i,0]), ", Batsman : ",index2batsmen.get(test1.iloc[i,1]), '- Times wicket Prediction:',model.predict(test1.iloc[[i]]))
1/1 [==============================] - 0s 90ms/step
Bowler : Harbhajan Singh , Batsman :  AM Nayar - Times wicket Prediction: [[1.0114906]]
1/1 [==============================] - 0s 18ms/step
Bowler : T Natarajan , Batsman :  Arshdeep Singh - Times wicket Prediction: [[0.98656166]]
1/1 [==============================] - 0s 19ms/step
Bowler : KK Ahmed , Batsman :  A Mishra - Times wicket Prediction: [[1.0504484]]
1/1 [==============================] - 0s 24ms/step
Bowler : M Muralitharan , Batsman :  F du Plessis - Times wicket Prediction: [[1.0941994]]
1/1 [==============================] - 0s 25ms/step
Bowler : SK Warne , Batsman :  DR Smith - Times wicket Prediction: [[1.0679393]]
1/1 [==============================] - 0s 28ms/step
Bowler : Mohammad Nabi , Batsman :  Ishan Kishan - Times wicket Prediction: [[1.403399]]
1/1 [==============================] - 0s 32ms/step
Bowler : R Bhatia , Batsman :  DJ Thornely - Times wicket Prediction: [[0.89399755]]
1/1 [==============================] - 0s 26ms/step
Bowler : SP Narine , Batsman :  MC Henriques - Times wicket Prediction: [[1.1997008]]
1/1 [==============================] - 0s 19ms/step
Bowler : AS Rajpoot , Batsman :  K Gowtham - Times wicket Prediction: [[0.9911405]]
1/1 [==============================] - 0s 21ms/step
Bowler : K Rabada , Batsman :  P Simran Singh - Times wicket Prediction: [[1.0064855]]

g) The embedding can be visualised using Google’s Embedding Projector, which identifies other batsmen who have similar characteristics. Here Cosine Similarity is used for grouping similar batsmen of IPL

The closest neighbor for AB De Villiers in IPL is SK Raina, then Rohit Sharma as seen in the visualisation below

B. Bowler Recommender T20 (BowlerRecommenderT20M1A.ipynb)

Similar to how batsman was set up,

The steps are

a) Upload data for T20 Batsman vs Bowler with Total runs scored. This will be a sparse matrix

b) Create integer dictionaries for batsman & bowler

d) Minimise loss for wicket taken

e) Display embeddings of bowlers using Tensorboard Embeddings Projector

Minimizing the loss for wicket taken using SGD

tf.random.set_seed(4)
vector_size=len(batsman2index)

#Normalize target variable
df4=df31[['bowler','batsman','totalRuns','fours','sixes','ballsFaced']]
df4['normalizedRuns'] = (df4['totalRuns'] -df4['totalRuns'].mean())/df4['totalRuns'].std()
print(df4)
train_dataset = df4.sample(frac=0.8,random_state=0)
test_dataset = df4.drop(train_dataset.index)
train_dataset1 = train_dataset[['batsman','bowler','fours','sixes','ballsFaced']]
test_dataset1 = test_dataset[['batsman','bowler','fours','sixes','ballsFaced']]

train_labels = train_dataset.pop('normalizedRuns')
test_labels = test_dataset.pop('normalizedRuns')
train_labels
print(train_dataset1)

# Create a Deep Learning model with keras
model = tf.keras.Sequential([
tf.keras.layers.Embedding(vector_size,16,input_length=5),
tf.keras.layers.Flatten(),
keras.layers.Dropout(.2),
keras.layers.Dense(16),

keras.layers.Dense(8,activation=tf.nn.relu),

keras.layers.Dense(4,activation=tf.nn.relu),
keras.layers.Dense(1)
])

# Print the model summary
#model.summary()
# Use the Adam optimizer with a learning rate of 0.01
#optimizer=keras.optimizers.RMSprop(learning_rate=0.001, rho=0.2, momentum=0.2, epsilon=1e-07)
#optimizer=keras.optimizers.SGD(learning_rate=.009,momentum=0.1) - Works without dropout
optimizer=keras.optimizers.SGD(learning_rate=.01,momentum=0.1)

model.compile(loss='mean_squared_error',
optimizer=optimizer,
)

# Setup the training parameters
#model.compile(loss='binary_crossentropy',optimizer='rmsprop',metrics=['accuracy'])
# Create a model
history=model.fit(
train_dataset1, train_labels,batch_size=32,
epochs=40, validation_data = (test_dataset1,test_labels), verbose=1)
model.predict(train_dataset1[1:10])
df5= df4[['batsman','bowler','fours','sixes','ballsFaced']]
test1 = df5.sample(n=10)
model.predict(test1)
#(model.predict(test1)* df4['totalRuns'].std()) + df4['totalRuns'].mean()
for i in range(test1.shape[0]):
print('Batsman :', index2batsman.get(test1.iloc[i,0]), ", Bowler : ",index2bowler.get(test1.iloc[i,1]), '- Total runs Prediction:',(model.predict(test1.iloc[i])* df4['totalRuns'].std()) + df4['totalRuns'].mean())
1/1 [==============================] - 0s 396ms/step
1/1 [==============================] - 0s 112ms/step
1/1 [==============================] - 0s 183ms/step
Batsman : G Chohan , Bowler :  Khawar Ali - Total runs Prediction: [[1.8883028]]
1/1 [==============================] - 0s 56ms/step
Batsman : Umar Akmal , Bowler :  LJ Wright - Total runs Prediction: [[9.305391]]
1/1 [==============================] - 0s 68ms/step
Batsman : M Shumba , Bowler :  Simi Singh - Total runs Prediction: [[19.662743]]
1/1 [==============================] - 0s 30ms/step
Batsman : CH Gayle , Bowler :  RJW Topley - Total runs Prediction: [[16.854687]]
1/1 [==============================] - 0s 39ms/step
Batsman : BA King , Bowler :  Taskin Ahmed - Total runs Prediction: [[3.5154686]]
1/1 [==============================] - 0s 102ms/step
Batsman : KD Shah , Bowler :  Avesh Khan - Total runs Prediction: [[8.411661]]
1/1 [==============================] - 0s 38ms/step
Batsman : ST Jayasuriya , Bowler :  SCJ Broad - Total runs Prediction: [[5.867449]]
1/1 [==============================] - 0s 45ms/step
Batsman : AB de Villiers , Bowler :  Saeed Ajmal - Total runs Prediction: [[15.150892]]
1/1 [==============================] - 0s 46ms/step
Batsman : SV Samson , Bowler :  J Little - Total runs Prediction: [[10.44426]]
1/1 [==============================] - 0s 102ms/step
Batsman : Zawar Farid , Bowler :  GJ Delany - Total runs Prediction: [[1.9770675]]

Identifying similar bowlers using Embeddings Projector for T20

Bhuvaneshwar Kumar’s performance is closest to CR Woakes

Note: Incidentally the accuracy in the above model was not too good. I may work on this again later!

C) Bowler Embeddings IPL – Grouping similar bowlers of IPL with Embeddings Projector (BowlerRecommenderIPLA.ipynb)

D) Batting Embeddings T20 – Grouping similar batsmen of T20 (BatsmanRecommenderT20MA.ipynb)

The Tensorboard Pmbeddings projector is also interesting. There are multiple ways the data can be visualised namely UMAP, T-SNE, PCA(included). You could play with it.

As mentioned above the Colab notebooks and data are available at Github embeddings

The ability to identify batsmen & bowlers who would perform similarly against specific bowling attacks coupled with the average runs & strike rate should give a good measure of a player’s performance.

Take a look at some of my other posts

To see all posts click Index of posts

# My book ‘Practical Machine Learning in R and Python: Third edition’ on Amazon

Are you wondering whether to get into the ‘R’ bus or ‘Python’ bus?
My suggestion is to you is “Why not get into the ‘R and Python’ train?”

The third edition of my book ‘Practical Machine Learning with R and Python – Machine Learning in stereo’ is now available in both paperback ($12.99) and kindle ($8.99/Rs449) versions.  In the third edition all code sections have been re-formatted to use the fixed width font ‘Consolas’. This neatly organizes output which have columns like confusion matrix, dataframes etc to be columnar, making the code more readable.  There is a science to formatting too!! which improves the look and feel. It is little wonder that Steve Jobs had a keen passion for calligraphy! Additionally some typos have been fixed.

In this book I implement some of the most common, but important Machine Learning algorithms in R and equivalent Python code.
1. Practical machine with R and Python: Third Edition – Machine Learning in Stereo(Paperback-$12.99) 2. Practical machine with R and Python Third Edition – Machine Learning in Stereo(Kindle-$8.99/Rs449)

This book is ideal both for beginners and the experts in R and/or Python. Those starting their journey into datascience and ML will find the first 3 chapters useful, as they touch upon the most important programming constructs in R and Python and also deal with equivalent statements in R and Python. Those who are expert in either of the languages, R or Python, will find the equivalent code ideal for brushing up on the other language. And finally,those who are proficient in both languages, can use the R and Python implementations to internalize the ML algorithms better.

Here is a look at the topics covered

Preface …………………………………………………………………………….4
Introduction ………………………………………………………………………6
1. Essential R ………………………………………………………………… 8
2. Essential Python for Datascience ……………………………………………57
3. R vs Python …………………………………………………………………81
4. Regression of a continuous variable ……………………………………….101
5. Classification and Cross Validation ………………………………………..121
6. Regression techniques and regularization ………………………………….146
7. SVMs, Decision Trees and Validation curves ………………………………191
8. Splines, GAMs, Random Forests and Boosting ……………………………222
9. PCA, K-Means and Hierarchical Clustering ………………………………258
References ……………………………………………………………………..269

Hope you have a great time learning as I did while implementing these algorithms!

# My book ‘Deep Learning from first principles:Second Edition’ now on Amazon

The second edition of my book ‘Deep Learning from first principles:Second Edition- In vectorized Python, R and Octave’, is now available on Amazon, in both paperback ($18.99) and kindle ($9.99/Rs449/-)  versions. Since this book is almost 70% code, all functions, and code snippets have been formatted to use the fixed-width font ‘Lucida Console’. In addition line numbers have been added to all code snippets. This makes the code more organized and much more readable. I have also fixed typos in the book

The book includes the following chapters

Table of Contents
Preface 4
Introduction 6
1. Logistic Regression as a Neural Network 8
2. Implementing a simple Neural Network 23
3. Building a L- Layer Deep Learning Network 48
4. Deep Learning network with the Softmax 85
5. MNIST classification with Softmax 103
6. Initialization, regularization in Deep Learning 121
7. Gradient Descent Optimization techniques 167
8. Gradient Check in Deep Learning 197
1. Appendix A 214
2. Appendix 1 – Logistic Regression as a Neural Network 220
3. Appendix 2 - Implementing a simple Neural Network 227
4. Appendix 3 - Building a L- Layer Deep Learning Network 240
5. Appendix 4 - Deep Learning network with the Softmax 259
6. Appendix 5 - MNIST classification with Softmax 269
7. Appendix 6 - Initialization, regularization in Deep Learning 302
8. Appendix 7 - Gradient Descent Optimization techniques 344
9. Appendix 8 – Gradient Check 405
References 475

To see posts click Index of Posts

# My book ‘Practical Machine Learning in R and Python: Second edition’ on Amazon

Note: The 3rd edition of this book is now available My book ‘Practical Machine Learning in R and Python: Third edition’ on Amazon

The third edition of my book ‘Practical Machine Learning with R and Python – Machine Learning in stereo’ is now available in both paperback ($12.99) and kindle ($9.99/Rs449) versions.  This second edition includes more content,  extensive comments and formatting for better readability.

In this book I implement some of the most common, but important Machine Learning algorithms in R and equivalent Python code.
1. Practical machine with R and Python: Third Edition – Machine Learning in Stereo(Paperback-$12.99) 2. Practical machine with R and Third Edition – Machine Learning in Stereo(Kindle-$9.99/Rs449)

This book is ideal both for beginners and the experts in R and/or Python. Those starting their journey into datascience and ML will find the first 3 chapters useful, as they touch upon the most important programming constructs in R and Python and also deal with equivalent statements in R and Python. Those who are expert in either of the languages, R or Python, will find the equivalent code ideal for brushing up on the other language. And finally,those who are proficient in both languages, can use the R and Python implementations to internalize the ML algorithms better.

Here is a look at the topics covered

Preface …………………………………………………………………………….4
Introduction ………………………………………………………………………6
1. Essential R ………………………………………………………………… 8
2. Essential Python for Datascience ……………………………………………57
3. R vs Python …………………………………………………………………81
4. Regression of a continuous variable ……………………………………….101
5. Classification and Cross Validation ………………………………………..121
6. Regression techniques and regularization ………………………………….146
7. SVMs, Decision Trees and Validation curves ………………………………191
8. Splines, GAMs, Random Forests and Boosting ……………………………222
9. PCA, K-Means and Hierarchical Clustering ………………………………258
References ……………………………………………………………………..269

Hope you have a great time learning as I did while implementing these algorithms!

# Deep Learning from first principles in Python, R and Octave – Part 6

“Today you are You, that is truer than true. There is no one alive who is Youer than You.”
Dr. Seuss

“Explanations exist; they have existed for all time; there is always a well-known solution to every human problem — neat, plausible, and wrong.”
H L Mencken

# Introduction

In this 6th instalment of ‘Deep Learning from first principles in Python, R and Octave-Part6’, I look at a couple of different initialization techniques used in Deep Learning, L2 regularization and the ‘dropout’ method. Specifically, I implement “He initialization” & “Xavier Initialization”. My earlier posts in this series of Deep Learning included

1. Part 1 – In the 1st part, I implemented logistic regression as a simple 2 layer Neural Network
2. Part 2 – In part 2, implemented the most basic of Neural Networks, with just 1 hidden layer, and any number of activation units in that hidden layer. The implementation was in vectorized Python, R and Octave
3. Part 3 -In part 3, I derive the equations and also implement a L-Layer Deep Learning network with either the relu, tanh or sigmoid activation function in Python, R and Octave. The output activation unit was a sigmoid function for logistic classification
4. Part 4 – This part looks at multi-class classification, and I derive the Jacobian of a Softmax function and implement a simple problem to perform multi-class classification.
5. Part 5 – In the 5th part, I extend the L-Layer Deep Learning network implemented in Part 3, to include the Softmax classification. I also use this L-layer implementation to classify MNIST handwritten digits with Python, R and Octave.

The code in Python, R and Octave are identical, and just take into account some of the minor idiosyncrasies of the individual language. In this post, I implement different initialization techniques (random, He, Xavier), L2 regularization and finally dropout. Hence my generic L-Layer Deep Learning network includes these additional enhancements for enabling/disabling initialization methods, regularization or dropout in the algorithm. It already included sigmoid & softmax output activation for binary and multi-class classification, besides allowing relu, tanh and sigmoid activation for hidden units.

A video presentation of regularization and initialization techniques can be also be viewed in Neural Networks 6

This R Markdown file and the code for Python, R and Octave can be cloned/downloaded from Github at DeepLearning-Part6

Checkout my book ‘Deep Learning from first principles: Second Edition – In vectorized Python, R and Octave’. My book starts with the implementation of a simple 2-layer Neural Network and works its way to a generic L-Layer Deep Learning Network, with all the bells and whistles. The derivations have been discussed in detail. The code has been extensively commented and included in its entirety in the Appendix sections. My book is available on Amazon as paperback ($18.99) and in kindle version($9.99/Rs449).

You may also like my companion book “Practical Machine Learning with R and Python:Second Edition- Machine Learning in stereo” available in Amazon in paperback($10.99) and Kindle($7.99/Rs449) versions. This book is ideal for a quick reference of the various ML functions and associated measurements in both R and Python which are essential to delve deep into Deep Learning.

## 1. Initialization techniques

The usual initialization technique is to generate Gaussian or uniform random numbers and multiply it by a small value like 0.01. Two techniques which are used to speed up convergence is the He initialization or Xavier. These initialization techniques enable gradient descent to converge faster.

## 1.1 a Default initialization – Python

This technique just initializes the weights to small random values based on Gaussian or uniform distribution

import numpy as np
import matplotlib
import matplotlib.pyplot as plt
import sklearn.linear_model
import pandas as pd
import sklearn
import sklearn.datasets
train_X, train_Y, test_X, test_Y = load_dataset()
# Set the layers dimensions
layersDimensions = [2,7,1]

# Train a deep learning network with random initialization
parameters = L_Layer_DeepModel(train_X, train_Y, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid",learningRate = 0.6, num_iterations = 9000, initType="default", print_cost = True,figure="fig1.png")

# Clear the plot
plt.clf()
plt.close()

# Plot the decision boundary
plot_decision_boundary(lambda x: predict(parameters, x.T), train_X, train_Y,str(0.6),figure1="fig2.png")

## 1.1 b He initialization – Python

‘He’ initialization attributed to He et al, multiplies the random weights by
$\sqrt{\frac{2}{dimension\ of\ previous\ layer}}$

import numpy as np
import matplotlib
import matplotlib.pyplot as plt
import sklearn.linear_model
import pandas as pd
import sklearn
import sklearn.datasets

train_X, train_Y, test_X, test_Y = load_dataset()
# Set the layers dimensions
layersDimensions = [2,7,1]

# Train a deep learning network with He  initialization
parameters = L_Layer_DeepModel(train_X, train_Y, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid", learningRate =0.6,    num_iterations = 10000,initType="He",print_cost = True,                           figure="fig3.png")

plt.clf()
plt.close()
# Plot the decision boundary
plot_decision_boundary(lambda x: predict(parameters, x.T), train_X, train_Y,str(0.6),figure1="fig4.png")

## 1.1 c Xavier initialization – Python

Xavier  initialization multiply the random weights by
$\sqrt{\frac{1}{dimension\ of\ previous\ layer}}$

import numpy as np
import matplotlib
import matplotlib.pyplot as plt
import sklearn.linear_model
import pandas as pd
import sklearn
import sklearn.datasets

train_X, train_Y, test_X, test_Y = load_dataset()
# Set the layers dimensions
layersDimensions = [2,7,1]

# Train a L layer Deep Learning network
parameters = L_Layer_DeepModel(train_X, train_Y, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid",
learningRate = 0.6,num_iterations = 10000, initType="Xavier",print_cost = True,
figure="fig5.png")

# Plot the decision boundary
plot_decision_boundary(lambda x: predict(parameters, x.T), train_X, train_Y,str(0.6),figure1="fig6.png")

## 1.2a Default initialization – R

source("DLfunctions61.R")
x <- z[,1:2]
y <- z[,3]
X <- t(x)
Y <- t(y)
#Set the layer dimensions
layersDimensions = c(2,11,1)
# Train a deep learning network
retvals = L_Layer_DeepModel(X, Y, layersDimensions,
hiddenActivationFunc='relu',
outputActivationFunc="sigmoid",
learningRate = 0.5,
numIterations = 8000,
initType="default",
print_cost = True)
#Plot the cost vs iterations
iterations <- seq(0,8000,1000)
costs=retvals$costs df=data.frame(iterations,costs) ggplot(df,aes(x=iterations,y=costs)) + geom_point() + geom_line(color="blue") + ggtitle("Costs vs iterations") + xlab("No of iterations") + ylab("Cost") # Plot the decision boundary plotDecisionBoundary(z,retvals,hiddenActivationFunc="relu",lr=0.5) ## 1.2b He initialization – R The code for ‘He’ initilaization in R is included below # He Initialization model for L layers # Input : List of units in each layer # Returns: Initial weights and biases matrices for all layers # He initilization multiplies the random numbers with sqrt(2/layerDimensions[previouslayer]) HeInitializeDeepModel <- function(layerDimensions){ set.seed(2) # Initialize empty list layerParams <- list() # Note the Weight matrix at layer 'l' is a matrix of size (l,l-1) # The Bias is a vectors of size (l,1) # Loop through the layer dimension from 1.. L # Indices in R start from 1 for(l in 2:length(layersDimensions)){ # Initialize a matrix of small random numbers of size l x l-1 # Create random numbers of size l x l-1 w=rnorm(layersDimensions[l]*layersDimensions[l-1]) # Create a weight matrix of size l x l-1 with this initial weights and # Add to list W1,W2... WL # He initialization - Divide by sqrt(2/layerDimensions[previous layer]) layerParams[[paste('W',l-1,sep="")]] = matrix(w,nrow=layersDimensions[l], ncol=layersDimensions[l-1])*sqrt(2/layersDimensions[l-1]) layerParams[[paste('b',l-1,sep="")]] = matrix(rep(0,layersDimensions[l]), nrow=layersDimensions[l],ncol=1) } return(layerParams) }  The code in R below uses He initialization to learn the data source("DLfunctions61.R") # Load the data z <- as.matrix(read.csv("circles.csv",header=FALSE)) x <- z[,1:2] y <- z[,3] X <- t(x) Y <- t(y) # Set the layer dimensions layersDimensions = c(2,11,1) # Train a deep learning network retvals = L_Layer_DeepModel(X, Y, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid", learningRate = 0.5, numIterations = 9000, initType="He", print_cost = True) #Plot the cost vs iterations iterations <- seq(0,9000,1000) costs=retvals$costs
df=data.frame(iterations,costs)
ggplot(df,aes(x=iterations,y=costs)) + geom_point() + geom_line(color="blue") +
ggtitle("Costs vs iterations") + xlab("No of iterations") + ylab("Cost")

# Plot the decision boundary
plotDecisionBoundary(z,retvals,hiddenActivationFunc="relu",0.5,lr=0.5)

## 1.2c Xavier initialization – R

## Xav initialization
# Set the layer dimensions
layersDimensions = c(2,11,1)
# Train a deep learning network
retvals = L_Layer_DeepModel(X, Y, layersDimensions,
hiddenActivationFunc='relu',
outputActivationFunc="sigmoid",
learningRate = 0.5,
numIterations = 9000,
initType="Xav",
print_cost = True)
#Plot the cost vs iterations
iterations <- seq(0,9000,1000)
costs=retvals$costs df=data.frame(iterations,costs) ggplot(df,aes(x=iterations,y=costs)) + geom_point() + geom_line(color="blue") + ggtitle("Costs vs iterations") + xlab("No of iterations") + ylab("Cost") # Plot the decision boundary plotDecisionBoundary(z,retvals,hiddenActivationFunc="relu",0.5) ## 1.3a Default initialization – Octave source("DL61functions.m") # Read the data data=csvread("circles.csv"); X=data(:,1:2); Y=data(:,3); # Set the layer dimensions layersDimensions = [2 11 1]; #tanh=-0.5(ok), #relu=0.1 best! # Train a deep learning network [weights biases costs]=L_Layer_DeepModel(X', Y', layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid", learningRate = 0.5, lambd=0, keep_prob=1, numIterations = 10000, initType="default"); # Plot cost vs iterations plotCostVsIterations(10000,costs) #Plot decision boundary plotDecisionBoundary(data,weights, biases,keep_prob=1, hiddenActivationFunc="relu")  ## 1.3b He initialization – Octave source("DL61functions.m") #Load data data=csvread("circles.csv"); X=data(:,1:2); Y=data(:,3); # Set the layer dimensions layersDimensions = [2 11 1]; #tanh=-0.5(ok), #relu=0.1 best! # Train a deep learning network [weights biases costs]=L_Layer_DeepModel(X', Y', layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid", learningRate = 0.5, lambd=0, keep_prob=1, numIterations = 8000, initType="He"); plotCostVsIterations(8000,costs) #Plot decision boundary plotDecisionBoundary(data,weights, biases,keep_prob=1,hiddenActivationFunc="relu")  ## 1.3c Xavier initialization – Octave The code snippet for Xavier initialization in Octave is shown below source("DL61functions.m") # Xavier Initialization for L layers # Input : List of units in each layer # Returns: Initial weights and biases matrices for all layers function [W b] = XavInitializeDeepModel(layerDimensions) rand ("seed", 3); # note the Weight matrix at layer 'l' is a matrix of size (l,l-1) # The Bias is a vectors of size (l,1) # Loop through the layer dimension from 1.. L # Create cell arrays for Weights and biases for l =2:size(layerDimensions)(2) W{l-1} = rand(layerDimensions(l),layerDimensions(l-1))* sqrt(1/layerDimensions(l-1)); # Multiply by .01 b{l-1} = zeros(layerDimensions(l),1); endfor end  The Octave code below uses Xavier initialization source("DL61functions.m") #Load data data=csvread("circles.csv"); X=data(:,1:2); Y=data(:,3); #Set layer dimensions layersDimensions = [2 11 1]; #tanh=-0.5(ok), #relu=0.1 best! # Train a deep learning network [weights biases costs]=L_Layer_DeepModel(X', Y', layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid", learningRate = 0.5, lambd=0, keep_prob=1, numIterations = 8000, initType="Xav"); plotCostVsIterations(8000,costs) plotDecisionBoundary(data,weights, biases,keep_prob=1,hiddenActivationFunc="relu")  ## 2.1a Regularization : Circles data – Python The cross entropy cost for Logistic classification is given as $J = \frac{1}{m}\sum_{i=1}^{m}y^{i}log((a^{L})^{(i)}) - (1-y^{i})log((a^{L})^{(i)})$ The regularized L2 cost is given by $J = \frac{1}{m}\sum_{i=1}^{m}y^{i}log((a^{L})^{(i)}) - (1-y^{i})log((a^{L})^{(i)}) + \frac{\lambda}{2m}\sum \sum \sum W_{kj}^{l}$ import numpy as np import matplotlib import matplotlib.pyplot as plt import sklearn.linear_model import pandas as pd import sklearn import sklearn.datasets exec(open("DLfunctions61.py").read()) #Load the data train_X, train_Y, test_X, test_Y = load_dataset() # Set the layers dimensions layersDimensions = [2,7,1] # Train a deep learning network parameters = L_Layer_DeepModel(train_X, train_Y, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid",learningRate = 0.6, lambd=0.1, num_iterations = 9000, initType="default", print_cost = True,figure="fig7.png") # Clear the plot plt.clf() plt.close() # Plot the decision boundary plot_decision_boundary(lambda x: predict(parameters, x.T), train_X, train_Y,str(0.6),figure1="fig8.png") plt.clf() plt.close() #Plot the decision boundary plot_decision_boundary(lambda x: predict(parameters, x.T,keep_prob=0.9), train_X, train_Y,str(2.2),"fig8.png",) ## 2.1 b Regularization: Spiral data – Python import numpy as np import matplotlib import matplotlib.pyplot as plt import sklearn.linear_model import pandas as pd import sklearn import sklearn.datasets exec(open("DLfunctions61.py").read()) N = 100 # number of points per class D = 2 # dimensionality K = 3 # number of classes X = np.zeros((N*K,D)) # data matrix (each row = single example) y = np.zeros(N*K, dtype='uint8') # class labels for j in range(K): ix = range(N*j,N*(j+1)) r = np.linspace(0.0,1,N) # radius t = np.linspace(j*4,(j+1)*4,N) + np.random.randn(N)*0.2 # theta X[ix] = np.c_[r*np.sin(t), r*np.cos(t)] y[ix] = j # Plot the data plt.scatter(X[:, 0], X[:, 1], c=y, s=40, cmap=plt.cm.Spectral) plt.clf() plt.close() #Set layer dimensions layersDimensions = [2,100,3] y1=y.reshape(-1,1).T # Train a deep learning network parameters = L_Layer_DeepModel(X.T, y1, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="softmax", learningRate = 1,lambd=1e-3, num_iterations = 5000, print_cost = True,figure="fig9.png") plt.clf() plt.close() W1=parameters['W1'] b1=parameters['b1'] W2=parameters['W2'] b2=parameters['b2'] plot_decision_boundary1(X, y1,W1,b1,W2,b2,figure2="fig10.png") ## 2.2a Regularization: Circles data – R source("DLfunctions61.R") #Load data df=read.csv("circles.csv",header=FALSE) z <- as.matrix(read.csv("circles.csv",header=FALSE)) x <- z[,1:2] y <- z[,3] X <- t(x) Y <- t(y) #Set layer dimensions layersDimensions = c(2,11,1) # Train a deep learning network retvals = L_Layer_DeepModel(X, Y, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid", learningRate = 0.5, lambd=0.1, numIterations = 9000, initType="default", print_cost = True)  #Plot the cost vs iterations iterations <- seq(0,9000,1000) costs=retvals$costs
df=data.frame(iterations,costs)
ggplot(df,aes(x=iterations,y=costs)) + geom_point() + geom_line(color="blue") +
ggtitle("Costs vs iterations") + xlab("No of iterations") + ylab("Cost")

# Plot the decision boundary
plotDecisionBoundary(z,retvals,hiddenActivationFunc="relu",0.5)

## 2.2b Regularization:Spiral data – R

# Read the spiral dataset
source("DLfunctions61.R")

# Setup the data
X <- Z[,1:2]
y <- Z[,3]
X <- t(X)
Y <- t(y)
layersDimensions = c(2, 100, 3)
# Train a deep learning network
retvals = L_Layer_DeepModel(X, Y, layersDimensions,
hiddenActivationFunc='relu',
outputActivationFunc="softmax",
learningRate = 0.5,
lambd=0.01,
numIterations = 9000,
print_cost = True)
print_cost = True)
parameters<-retvals$parameters plotDecisionBoundary1(Z,parameters) 2.3a Regularization: Circles data – Octave source("DL61functions.m") #Load data data=csvread("circles.csv"); X=data(:,1:2); Y=data(:,3); layersDimensions = [2 11 1]; #tanh=-0.5(ok), #relu=0.1 best! # Train a deep learning network [weights biases costs]=L_Layer_DeepModel(X', Y', layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid", learningRate = 0.5, lambd=0.2, keep_prob=1, numIterations = 8000, initType="default"); plotCostVsIterations(8000,costs) #Plot decision boundary plotDecisionBoundary(data,weights, biases,keep_prob=1,hiddenActivationFunc="relu")  ## 2.3b Regularization:Spiral data 2 – Octave source("DL61functions.m") data=csvread("spiral.csv"); # Setup the data X=data(:,1:2); Y=data(:,3); layersDimensions = [2 100 3] # Train a deep learning network [weights biases costs]=L_Layer_DeepModel(X', Y', layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="softmax", learningRate = 0.6, lambd=0.2, keep_prob=1, numIterations = 10000); plotCostVsIterations(10000,costs) #Plot decision boundary plotDecisionBoundary1(data,weights, biases,keep_prob=1,hiddenActivationFunc="relu")  ## 3.1 a Dropout: Circles data – Python The ‘dropout’ regularization technique was used with great effectiveness, to prevent overfitting by Alex Krizhevsky, Ilya Sutskever and Prof Geoffrey E. Hinton in the Imagenet classification with Deep Convolutional Neural Networks The technique of dropout works by dropping a random set of activation units in each hidden layer, based on a ‘keep_prob’ criteria in the forward propagation cycle. Here is the code for Octave. A ‘dropoutMat’ is created for each layer which specifies which units to drop Note: The same ‘dropoutMat has to be used which computing the gradients in the backward propagation cycle. Hence the dropout matrices are stored in a cell array.  for l =1:L-1 ... D=rand(size(A)(1),size(A)(2)); D = (D < keep_prob) ; # Zero out some hidden units A= A .* D; # Divide by keep_prob to keep the expected value of A the same A = A ./ keep_prob; # Store D in a dropoutMat cell array dropoutMat{l}=D; ... endfor In the backward propagation cycle we have  for l =(L-1):-1:1 ... D = dropoutMat{l}; # Zero out the dAl based on same dropout matrix dAl= dAl .* D; # Divide by keep_prob to maintain the expected value dAl = dAl ./ keep_prob; ... endfor  import numpy as np import matplotlib import matplotlib.pyplot as plt import sklearn.linear_model import pandas as pd import sklearn import sklearn.datasets exec(open("DLfunctions61.py").read()) #Load the data train_X, train_Y, test_X, test_Y = load_dataset() # Set the layers dimensions layersDimensions = [2,7,1] # Train a deep learning network parameters = L_Layer_DeepModel(train_X, train_Y, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid",learningRate = 0.6, keep_prob=0.7, num_iterations = 9000, initType="default", print_cost = True,figure="fig11.png") # Clear the plot plt.clf() plt.close() # Plot the decision boundary plot_decision_boundary(lambda x: predict(parameters, x.T,keep_prob=0.7), train_X, train_Y,str(0.6),figure1="fig12.png")  ### 3.1b Dropout: Spiral data – Python import numpy as np import matplotlib import matplotlib.pyplot as plt import sklearn.linear_model import pandas as pd import sklearn import sklearn.datasets exec(open("DLfunctions61.py").read()) # Create an input data set - Taken from CS231n Convolutional Neural networks, # http://cs231n.github.io/neural-networks-case-study/ N = 100 # number of points per class D = 2 # dimensionality K = 3 # number of classes X = np.zeros((N*K,D)) # data matrix (each row = single example) y = np.zeros(N*K, dtype='uint8') # class labels for j in range(K): ix = range(N*j,N*(j+1)) r = np.linspace(0.0,1,N) # radius t = np.linspace(j*4,(j+1)*4,N) + np.random.randn(N)*0.2 # theta X[ix] = np.c_[r*np.sin(t), r*np.cos(t)] y[ix] = j # Plot the data plt.scatter(X[:, 0], X[:, 1], c=y, s=40, cmap=plt.cm.Spectral) plt.clf() plt.close() layersDimensions = [2,100,3] y1=y.reshape(-1,1).T # Train a deep learning network parameters = L_Layer_DeepModel(X.T, y1, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="softmax", learningRate = 1,keep_prob=0.9, num_iterations = 5000, print_cost = True,figure="fig13.png") plt.clf() plt.close() W1=parameters['W1'] b1=parameters['b1'] W2=parameters['W2'] b2=parameters['b2'] #Plot decision boundary plot_decision_boundary1(X, y1,W1,b1,W2,b2,figure2="fig14.png") ## 3.2a Dropout: Circles data – R source("DLfunctions61.R") #Load data df=read.csv("circles.csv",header=FALSE) z <- as.matrix(read.csv("circles.csv",header=FALSE)) x <- z[,1:2] y <- z[,3] X <- t(x) Y <- t(y) layersDimensions = c(2,11,1) # Train a deep learning network retvals = L_Layer_DeepModel(X, Y, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="sigmoid", learningRate = 0.5, keep_prob=0.8, numIterations = 9000, initType="default", print_cost = True) # Plot the decision boundary plotDecisionBoundary(z,retvals,keep_prob=0.6, hiddenActivationFunc="relu",0.5) ## 3.2b Dropout: Spiral data – R # Read the spiral dataset source("DLfunctions61.R") # Load data Z <- as.matrix(read.csv("spiral.csv",header=FALSE)) # Setup the data X <- Z[,1:2] y <- Z[,3] X <- t(X) Y <- t(y) # Train a deep learning network retvals = L_Layer_DeepModel(X, Y, layersDimensions, hiddenActivationFunc='relu', outputActivationFunc="softmax", learningRate = 0.1, keep_prob=0.90, numIterations = 9000, print_cost = True)  parameters<-retvals$parameters
#Plot decision boundary
plotDecisionBoundary1(Z,parameters)

## 3.3a Dropout: Circles data – Octave

data=csvread("circles.csv");

X=data(:,1:2);
Y=data(:,3);
layersDimensions = [2 11  1]; #tanh=-0.5(ok), #relu=0.1 best!

# Train a deep learning network
[weights biases costs]=L_Layer_DeepModel(X', Y', layersDimensions,
hiddenActivationFunc='relu',
outputActivationFunc="sigmoid",
learningRate = 0.5,
lambd=0,
keep_prob=0.8,
numIterations = 10000,
initType="default");
plotCostVsIterations(10000,costs)
#Plot decision boundary
plotDecisionBoundary1(data,weights, biases,keep_prob=1, hiddenActivationFunc="relu")


## 3.3b Dropout  Spiral data – Octave

source("DL61functions.m")

# Setup the data
X=data(:,1:2);
Y=data(:,3);

layersDimensions = [numFeats numHidden  numOutput];
# Train a deep learning network
[weights biases costs]=L_Layer_DeepModel(X', Y', layersDimensions,
hiddenActivationFunc='relu',
outputActivationFunc="softmax",
learningRate = 0.1,
lambd=0,
keep_prob=0.8,
numIterations = 10000);

plotCostVsIterations(10000,costs)
#Plot decision boundary
plotDecisionBoundary1(data,weights, biases,keep_prob=1, hiddenActivationFunc="relu")


Note: The Python, R and Octave code can be cloned/downloaded from Github at DeepLearning-Part6
Conclusion
This post further enhances my earlier L-Layer generic implementation of a Deep Learning network to include options for initialization techniques, L2 regularization or dropout regularization

To see all posts click Index of posts

# Presentation on ‘Machine Learning in plain English – Part 2’

This presentation is a continuation of my earlier presentation Presentation on ‘Machine Learning in plain English – Part 1’. As the title suggests, the presentation is devoid of any math or programming constructs, and just focuses on the concepts and approaches to different Machine Learning algorithms. In this 2nd part, I discuss KNN regression, KNN classification, Cross Validation techniques like (LOOCV, K-Fold)   feature selection methods including best-fit,forward-fit and backward fit and finally Ridge (L2) and Lasso Regression (L1)

If you would like to see the implementations of the discussed algorithms, in this presentation, do check out my book   My book ‘Practical Machine Learning with R and Python’ on Amazon

To see all post click Index of posts

# My book ‘Practical Machine Learning with R and Python’ on Amazon

Note: The 3rd edition of this book is now available My book ‘Practical Machine Learning in R and Python: Third edition’ on Amazon

My book ‘Practical Machine Learning with R and Python: Second Edition – Machine Learning in stereo’ is now available in both paperback ($10.99) and kindle ($7.99/Rs449) versions. In this book I implement some of the most common, but important Machine Learning algorithms in R and equivalent Python code. This is almost like listening to parallel channels of music in stereo!
1. Practical machine with R and Python: Third Edition – Machine Learning in Stereo(Paperback-$12.99) 2. Practical machine with R and Python Third Edition – Machine Learning in Stereo(Kindle-$8.99/Rs449)
This book is ideal both for beginners and the experts in R and/or Python. Those starting their journey into datascience and ML will find the first 3 chapters useful, as they touch upon the most important programming constructs in R and Python and also deal with equivalent statements in R and Python. Those who are expert in either of the languages, R or Python, will find the equivalent code ideal for brushing up on the other language. And finally,those who are proficient in both languages, can use the R and Python implementations to internalize the ML algorithms better.

Here is a look at the topics covered

Essential R …………………………………….. 7
Essential Python for Datascience ………………..   54
R vs Python ……………………………………. 77
Regression of a continuous variable ………………. 96
Classification and Cross Validation ……………….113
Regression techniques and regularization …………. 134
SVMs, Decision Trees and Validation curves …………175
Splines, GAMs, Random Forests and Boosting …………202
PCA, K-Means and Hierarchical Clustering …………. 234

Hope you have a great time learning as I did while implementing these algorithms!

# Simplifying Machine Learning: Bias, Variance, regularization and odd facts – Part 4

In both linear and logistic regression the choice of the degree of the polynomial for the hypothesis function is extremely critical. A low degree for the polynomial can result in an underfit, while a very high degree can overfit the data as shown below

The figure on the left the data is underfit as we try to fit the data with a first order polynomial which is a straight line. This is a case of strong ‘bias’

The rightmost figure a much higher polynomial is used. All the data points are covered by the polynomial curve however it is not effective in predicting other values. This is a case of overfitting or a high variance.

The middle figure is just right as it intuitively fits the data points the best possible way.

A similar problem exists with logistic regression as shown below

There are 2 ways to handle overfitting

a)      Reducing the number of features selected

b)      Using regularization

In regularization the magnitude of the parameters Ɵ is decreased to reduce the effect of overfitting

Hence if we choose a hypothesis function

hƟ (x) = Ɵ0 + Ɵ1x12 + Ɵ2x22 + Ɵ3x33 +  Ɵ4x44

The cost function for this without regularization as mentioned in earlier posts

J(Ɵ) = 1/2m Σ(hƟ (xi  – yi)2

Where the key is minimize the above function for the least error

The cost function with regularization becomes

J(Ɵ) = 1/2m Σ(hƟ (xi  – yi)2 + λ Σ Ɵj2

As can be seen the regularization now adds a factor Ɵj2  as a part of the cost function which needs to be minimized.

Hence with the regularization factor the problem of underfitting/overfitting can be solved

However the trick is determine the value of λ. If λ is too big then it would result in underfitting or resulting in a high bias.

Similarly the regularized equation for logistic regression is as shown below

J(Ɵ) = |1/m Σ  -y * log(hƟ (x))  – (1-y) * (log(1 – hƟ (x))  | + λ/2m Σ Ɵj2

Some tips suggested by Prof Andrew Ng while determining the parameters and features for regression

a)      Get as many training examples. It is worth spending more effort in getting as much examples