TensorFlow Intro

TensorFlow Intro.

What is TensorFlow?

The shortest definition would be, TensorFlow is a general-purpose library for graph-based computation.

But there is a variety of other ways to define TensorFlow, for example, Rodolfo Bonnin in his book – Building Machine Learning Projects with TensorFlow brings up definition like this:

“TensorFlow is an open source software library for numerical computation using data flow graphs. Nodes in the graph represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) passed between them.”

To quote the TensorFlow website, TensorFlow is an “open source software library for numerical computation using data flow graphs”.  Name TensorFlow derives from the operations which neural networks perform on multidimensional data arrays, often referred to as ‘tensors’. It is using data flow graphs and is capable of building and training variety of different machine learning algorithms including deep neural networks, at the same time, it is general enough to be applicable in a wide variety of other domains as well. Flexible architecture allows deploying computation to one or more CPUs or GPU in a desktop, server, or mobile device with a single API.

TensorFlow is Google Brain’s second generation machine learning system, released as open source software in 2015. TensorFlow is available on 64-bit Linux, macOS, and mobile computing platforms including Android and iOS. TensorFlow provides a Python API, as well as C++, Haskell, Java and Go APIs. Google’s machine learning framework became lately ‘hottest’ in data science world, it is particularly useful for building deep learning systems for predictive models involving natural language processing, audio, and images.

 

What is ‘Graph’ or ‘Data Flow Graph’? What is TensorFlow Session?

 

Trying to define what TensorFlow is, it is hard to avoid using word ‘graph’, or ‘data flow graph’, so what is that? The shortest definition would be, TensorFlow Graph is a description of computations. Deep learning (neural networks with many layers) uses mostly very simple mathematical operations – just many of them, on high dimensional data structures(tensors). Neural networks can have thousands or even millions of weights. Computing them, by interpreting every step (Python) would take forever.

That’s why we create a graph made up of defined tensors and mathematical operations and even initial values for variables. Only after we’ve created this ‘recipe’ we can pass it to what TensorFlow calls a session. To compute anything, a graph must be launched in a Session. The session runs the graph using very efficient and optimized code. Not only that, but many of the operations, such as matrix multiplication, are ones that can be parallelised by supported GPU (Graphics Processing Unit) and the session will do that for you. Also, TensorFlow is built to be able to distribute the processing across multiple machines and/or GPUs.

TensorFlow programs are usually divided into a construction phase, that assembles a graph, and an execution phase that uses a session to execute operations in the graph. To do machine learning in TensorFlow, you want to create tensors, adding operations (that output other tensors), and then executing the computation (running the computational graph). In particular, it’s important to realize that when you add an operation on tensors, it doesn’t execute immediately. TensorFlow waits for you to define all the operations you want to perform and then optimizes the computation graph, ‘deciding’ how to execute the computation, before generating the data. Because of this, tensors in TensorFlow are not so much holding the data as a placeholder for holding the data, waiting for the data to arrive when a computation is executed.

 

Prerequisites:

 

NEURAL NETWORKS – basics.

Before we move on to create our first model in TensorFlow, we’ll need to get the basics right, talk a bit about the structure of a simple neural network.

A simple neural network has some input units where the input goes. It also has hidden units, so-called because from a user’s perspective they’re hidden. And there are output units, from which we get the results. Off to the side are also bias units, which are there to help control the values emitted from the hidden and output units. Connecting all of these units are a bunch of weights, which are just numbers, each of which is associated with two units. The way we train neural network is to assign values to all those weights. That’s what training a neural network does, find suitable values for those weights. One step in “running” the neural network is to multiply the value of each weight by the value of its input unit, and then to store the result in the associated unit.

There is plenty of resources available online to get more background on the neural networks architectures, few examples below:

 

MATHEMATICS

Deep learning uses very simple mathematical operations, it would be recommended to get/refresh at least basics of them.  I recommend starting from one of the following:

 

PYTHON

It would be advised to have basics Python programming before moving forward, few available resources:

 

Let’s do it… TensorFlow first example code.

 

To keep things simple let’s start with ‘Halo World’ example.

importing TensorFlow

 

import tensorflow as tf

Declaring constants/variables, TensorFlow constants can be declared using the tf.constant function, and variables with the tf.Variable function.  The first element in both is the value to be assigned the constant/variable when it is initialised.  TensorFlow will infer the type of the constant/variable initialised value, but it can also be set explicitly using the optional dtype argument. It’s important to note that, as the Python code runs through these commands, the variables haven’t actually been declared as they would have been if you just had a standard Python declaration.

x = tf.constant(2.0) 
y = tf.Variable(3.0)

Lets make our code compute something, simple multiplication.

z = y * x

Now comes the time when we would like to see the outcome, except nothing, has been computed yet… welcome to the TensorFlow. To make use of TensorFlow variables and perform calculations, Session must be created and all variables must be initialized. We can do it using the following statements.

sess = tf.Session()

init = tf.global_variables_initializer()

sess.run(init)

 

We have Session and even all constants/variables in place. Let’s see the outcome.

print("z = y * x = ", sess.run(z))

 

If you see something like this:
‘z = y * x = 6.0’
Congratulations, you have just coded you first TensorFlow ‘model’.

Below whole code in one piece:

import tensorflow as tf
x = tf.constant(2.0)
y = tf.Variable(3.0)
z = y * x
sess = tf.Session()
init = tf.global_variables_initializer()
sess.run(init)
print("z = y * x = ", sess.run(z))

This tutorial, of course, will not end up like this and will be continued soon… in next part, we will code our first neural network in TensorFlow.

If You liked this post please share it on your social media, if you have any questions or comments please make use of contact form.

Recommended reading list below:

“MUST KNOW” from Python-Pandas for Data Science

“MUST KNOW” from Python-Pandas for Data Science.

Pandas is very popular Python library for data analysis, manipulation, and visualization, I would like to share my personal view on the list of most often used functions/snippets for data analysis.

1.Import Pandas to Python

import pandas as pd

2. Import data from CSV/Excel file

df=pd.read_csv('C:/Folder/mlhype.csv')   #imports whole csv to pd dataframe
df = pd.read_csv('C:/Folder/mlhype.csv', usecols=['abv', 'ibu'])  #imports selected columns
df = pd.read_excel('C:/Folder/mlhype.xlsx')  #imports excel file

3. Save data to CSV/Excel

df.to_csv('C:/Folder/mlhype.csv') #saves data frame to csv
df.to_excel('C:/Folder/mlhype.xlsx') #saves data frame to excel

4. Exploring data

df.head(5) #returns top 5 rows of data
df.tail(5) #returns bottom 5 rows of data
df.sample(5) #returns random 5 rows of data
df.shape #returns number of rows and columns
df.info() #returns index,data types, memory information
df.describe() #returns basic statistical summary of columns

5. Basic statistical functions

df.mean() #returns mean of columns
df.corr() #returns correlation table
df.count() #returns count of non-null's in column
df.max() #returns max value in each column
df.min() #returns min value in each column
df.median() #returns median of each colun
df.std() #returns standard deviation of each column

6. Selecting subsets

df['ColumnName'] #returns column 'ColumnName'
df[['ColumnName1','ColumnName2']] #returns multiple columns from the list
df.iloc[2,:] #selection by position - whole second row
df.iloc[:,2] #selection by position - whole second column
df.loc[:10,'ColumnName'] #returns first 11 rows of column
df.ix[2,'ColumnName'] #returns second element of column

7. Data cleansing

df.columns = ['a','b','c','d','e','f','g','h'] #rename column names
df.dropna() #drops all rows that contain missing values
df.fillna(0) #replaces missing values with 0 (or any other passed value)
df.fillna(df.mean()) #replaces missing values with mean(or any other passed function)

8.Filtering/sorting

df[df['ColumnName'] > 0.08] #returns rows with meeting criterion 
df[(df['ColumnName1']>2004) & (df['ColumnName2']==9)] #returns rows meeting multiple criteria
df.sort_values('ColumnName') #sorts by column in ascending order
df.sort_values('ColumnName',ascending=False) #sort by column in descending order

9. Data frames concatenation

pd.concat([DateFrame1, DataFrame2],axis=0) #concatenate rows vertically
pd.concat([DateFrame1, DataFrame2],axis=1) #concatenate rows horizontally

10.Adding new columns

df['NewColumn'] = 50 #creates new column with value 50 in each row
df['NewColumn3'] = df['NewColumn1']+df['NewColumn2'] #new column with value equal to sum of other columns
del df['NewColumn'] #deletes column

I hope you will find above useful, if you need more information on pandas, I recommend going to Pandas documentation or getting one of these books:

Numerai deep learning example.

In a previous post on Numerai, I have described very basic code to get into a world of machine learning competitions. This one will be a continuation, so if you haven’t read it I recommend to do it- here. In this post, we will add little more complexity to the whole process. We will split out 20% of training data as validation set so we can train different models and compare performance. And we will dive into deep neural nets as predicting model.

Ok, let’s do some machine learning…

Let’s start with importing what will be required, this step is similar to what we have done in the first model. Apart from Pandas, we import “StandardScaler” to preprocess data before feeding them into neural net. We will use “train_test_split” to split out 20% of data as a test set. “roc_auc_score” is a useful metric to check and compare performance of the model, we will also need neural net itself – that will be classifier from ‘scikit-neuralnetwork’ (sknn).

Imports first:

import pandas as pd
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from sklearn.metrics import roc_auc_score
from sknn.mlp import Classifier, Layer

As we have all required imports, we can load the data from csv(remember to update the system path to downloaded files):

train = pd.read_csv("/home/m/Numerai/numerai_datasets/numerai_training_data.csv")
test = pd.read_csv("/home/m/Numerai/numerai_datasets/numerai_tournament_data.csv")
sub = pd.read_csv("/home/m/Numerai/numerai_datasets/example_predictions.csv")

Some basic data manipulation required:

sub["t_id"]=test["t_id"]
test.drop("t_id", axis=1,inplace=True)
labels=train["target"]
train.drop("target", axis=1,inplace=True)
train=train.values
labels=labels.values

In next four lines, we will do what is called standardization. The result of standardization (or Z-score normalization) is that the features will be rescaled so that they’ll have the properties of a standard normal distribution with μ=0 and σ=1.

scaler = StandardScaler()
scaler.fit(train)
train = scaler.transform(train)
test = scaler.transform(test)

Next line of code will split original downloaded train set to train and test set, basically we set aside 20% of original train data to make sure we can check the out of the sample performance – to avoid overfitting.

X_train, X_test, y_train, y_test = train_test_split(train,labels, test_size=0.2, random_state=35)

Having all data preprocessed we are ready to define model, set number of layers in neural network, and a number of neurons in each layer. Below few lines of code to do it:

nn = Classifier(
layers=[
Layer("Tanh", units=50),
Layer("Tanh", units=200),
Layer("Tanh", units=200),
Layer("Tanh", units=50),
Layer("Softmax")],
learning_rule='adadelta',
learning_rate=0.01,
n_iter=5,
verbose=1,
loss_type='mcc')

“units=50” – states a number of neurons in each layer, number of neurons in first layer is determined by a number of features in data we will feed in.

“Tahn” – this is kind of activation function, you can use other as well eg. rectifier, expLin, sigmoid, or convolution. In last layer the activation function is Softmax – that’s usual output layer function for classification tasks. In our network we have five layers with a different number of neurons, there are no strict rules about number of neurons and layers so it is more art than science, you just need to try different versions and check what works best.

In our network we have five layers with a different number of neurons, there are no strict rules about a number of neurons and layers so it is more art than science, you just need to try different versions and check what works best. After layers we set learning rule to ‘adadelta’ again more choice available: sgd, momentum, nesterov, adagrad or rmsprop just try and check what works best.

“learning_rule=’adadelta'” – sets learning algorithm to ‘adadelta’, more choice available: sgd, momentum, nesterov, adagrad or rmsprop just try and check what works best, you can mix them for different layers.

“learning_rate=0.01” – learning rate, often as rule of thumb you start with ‘default’ value of 0.01, but other values can be used, mostly anything from 0.001 to 0.1.

“n_iter=5” – number of iterations ‘epochs’, the higher the number the longer process of learning will take, 5 is as example only, one need to look at error after each epoch, at some point it will stop dropping, I have seen anything from 50 to 5000 so feel free to play with it.

“verbose=1” – this parameter will let us see progress on screen.

“loss_type=’mcc’ ” – loss function, ‘mcc’ typical for classification tasks.

As the model is set, we can feed data and train it, depending on how powerful your pc is it can take from seconds to days. It is recommended to use GPU computing for neural networks training.

nn.fit(X_train, y_train)

Below line validates the model against 20% of data we have set aside before.

print('Overall AUC:', roc_auc_score(y_test, nn.predict_proba(X_test)[:,1]))

Using above code we can play around with different settings and neural networks architectures, checking the performance. After finding the best settings, they can be applied for prediction to be uploaded to Numerai, just run last three lines(just remember to update system path to save the file):

y_pred = nn.predict_proba(test)
sub["probability"]=y_pred[:,1]
sub.to_csv("/home/m/Numerai/numerai_datasets/Prediction.csv", index=False)

I hope above text was useful and you can now start playing around with deep learning for trading predictions for Numerai. If you have any comments or questions please feel free to contact me.

Full code below:

import pandas as pd
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from sklearn.metrics import roc_auc_score
from sknn.mlp import Classifier, Layer

train = pd.read_csv("/home/m/Numerai/numerai_datasets/numerai_training_data.csv")
test = pd.read_csv("/home/m/Numerai/numerai_datasets/numerai_tournament_data.csv")
sub = pd.read_csv("/home/m/Numerai/numerai_datasets/example_predictions.csv")

sub["t_id"]=test["t_id"]
test.drop("t_id", axis=1,inplace=True)

labels=train["target"]
train.drop("target", axis=1,inplace=True)

train=train.values
labels=labels.values

scaler = StandardScaler()
scaler.fit(train)
train = scaler.transform(train)
test = scaler.transform(test)

X_train, X_test, y_train, y_test = train_test_split(train,labels, test_size=0.2, random_state=35)

nn = Classifier(
layers=[
Layer("Tanh", units=50),
Layer("Tanh", units=200),
Layer("Tanh", units=200),
Layer("Tanh", units=50),
Layer("Softmax")],
learning_rule='adadelta',
learning_rate=0.01,
n_iter=5,
verbose=1,
loss_type='mcc')

nn.fit(X_train, y_train)

print('Overall AUC:', roc_auc_score(y_test, nn.predict_proba(X_test)[:,1]))

y_pred = nn.predict_proba(test)
sub["probability"]=y_pred[:,1]
sub.to_csv("/home/m/Numerai/numerai_datasets/Prediction.csv", index=False)

Machine learning competitions – Numerai example code.

In this post, I want to share, how simple it is to start competing in machine learning tournaments like Numerai. I will go step by step, line by line explaining what is doing what and why it is required.

Numerai is a global artificial intelligence competition in predicting financial markets. Numerai is a little bit similar to Kaggle but with clean datasets, so we can pass over long data cleansing process.  You just download the data, build a model, and upload your predictions, that’s it. To extract most of the data you would initially do some feature engineering, but for simplicity of this intro, we will pass this bit over.  One more thing we will pass on is splitting out validation set, the main aim of this exercise is to fit ‘machine learning’ model to training dataset. Later using fitted model, generate a prediction.  All together it shouldn’t take more than 14 simple lines of python code, you can run them as one piece or run part by part in interactive mode.

Let’s go, let’s do some machine learning…

A first thing to do is to go to numer.ai, click on ‘Download Training Data’  and download datasets, after unzipping the archive, you will have few files in there, we are interested mainly in three of them. It is worth noting what is a path to the folder as we will need it later.

I assume you have installed python and required libraries, if not there is plenty of online tutorials on how to do it, I recommend installing Anaconda distribution. It it time to open whatever IDE you use, and start coding, first few lines will be just importing what we will use later, that is Pandas and ScikitLearn.

import pandas as pd 
from sklearn.ensemble import GradientBoostingClassifier

Pandas is used to import data from csv files and do some basic data manipulations, GradientBoostingClassifier as part of ScikitLearn will be the model we will use to fit and do predict. As we have required libraries imported let’s use them… in next three lines, we will import data from csv to memory.  We will use ‘read_csv’  method from pandas, all you need to do is amend the full path to each file, wherever you have extracted numerai_datasets.zip.

train = pd.read_csv("/home/m/Numerai/numerai_datasets/numerai_training_data.csv")
test  = pd.read_csv("/home/m/Numerai/numerai_datasets/numerai_tournament_data.csv")   
sub  = pd.read_csv("/home/m/Numerai/numerai_datasets/example_predictions.csv")

What above code does it creates three data frames and imports the csv files we have we have previously extracted from downloaded numerai_datasets.zip.

‘train’ –  this dataset contains all required data to train our model, so it has both ‘features’ and ‘labels’, so you can say it has both questions and answers that our model will ‘learn’

‘test’ – this one contains features but does not contain ‘labels’, you can say it contains questions and our model will deliver answers.

‘sub’ – it is just template for uploading our prediction

Let’s move on,  in next line will copy all unique row id’s from ‘test’ to ‘sub’ to make sure each predicted value will be assigned to a right set of features, let’s say we put question number next to our answer so whoever checks the test would now.

sub["t_id"]=test["t_id"]

As we have copied the ids to ‘sub’, we don’t need them anymore in ‘test’ (all rows will stay in same order), so we can get rid of them.

test.drop("t_id", axis=1,inplace=True)

In next two lines, we will separate ‘labels’ or target values from train dataset.

labels=train["target"]
train.drop("target", axis=1,inplace=True)

As we have prepared ‘train’ dataset, we can get our model to learn from it. First, we select model we want to use, it will be Gradient BoostingClassifier from ScikitLearn – no specific reason for using this one, you can use whatever you like eg. random forest, linear regression…

grd = GradientBoostingClassifier()

As we have a model defined, let’s have it learn from ‘train’ data.

grd.fit(train,labels)

Ok, now our model is well trained and ready to make predictions, as the task is called ‘classification’ we will predict what is a probability of each set of features belongs to one of two classes ‘0’ or ‘1’.

y_pred = grd.predict_proba(test)

We have a long list of predicted probabilities called ‘y_pred’, let’s attach it to ‘id’ we had separated previously.

sub["probability"]=y_pred[:,1]

And save it in csv format, to get uploaded.

sub.to_csv("/home/m/Numerai/numerai_datasets/SimplePrediction.csv", index=False)

The last thing to do is go back to numer.ai website and click on ‘Upload Predictions’… Good luck.

This was very simplistic and introductory example to start playing with numer.ai competitions and machine learning. I will try and come back with gradually more complicated versions, if you have any questions, suggestions or comments please go to ‘About’ section and contact me directly.

The full code below:

import pandas as pd 
from sklearn.ensemble import GradientBoostingClassifier 
train = pd.read_csv("C:/Users/Downloads/numerai_datasets/numerai_training_data.csv") 
test = pd.read_csv("C:/Users/Downloads/numerai_datasets/numerai_tournament_data.csv") 
sub = pd.read_csv("C:/Users/Downloads/numerai_datasets/example_predictions.csv") 
sub["t_id"]=test["t_id"] 
test.drop("t_id", axis=1,inplace=True) 
labels=train["target"] 
train.drop("target", axis=1,inplace=True)
grd = GradientBoostingClassifier() 
grd.fit(train,labels) 
y_pred = grd.predict_proba(test) 
sub["probability"]=y_pred[:,1] 
sub.to_csv("C:/Users/Downloads/numerai_datasets/SimplePrediction.csv", index=False)