Your cart is empty.
Keep shopping
Teach on Udemy
Turn what you know into an opportunity and reach millions around the world.
Learn MoreMachine Learning with Python, scikit-learn and TensorFlow
Rating: 3.6 out of 53.6(27 ratings)
215 students
Machine Learning with Python, scikit-learn and TensorFlow
Apply Machine Learning techniques to solve real-world problems with Python, scikit-learn and TensorFlow
Created byPackt Publishing
Last updated 6/2018
English
What you'll learn
- Solve interesting, real-world problems using machine learning with Python
- Evaluate the performance of machine learning systems in common tasks
- Create pipelines to deal with real-world input data
- Traverse from concept to a production-ready machine learning setup/pipeline capable of real-world usage
- Use Python to visualize data spread across multiple dimensions and extract useful features to implement machine learning classification and regression algorithms from scratch in Python
- Predict the values of continuous variables using linear regression and K Nearest Neighbors to classify documents and images using logistic regression and support vector machines
Explore related topics
Course content
3 sections • 111 lectures • 9h 25m total length
- The Course Overview5:03
This video gives a glimpse of what you will learn through this video.
- Introduction to Machine Learning6:42
This video, will let you kick off your Python and machine learning journey with the basic, yet important concepts of machine learning.
- Get a high level overview of Machine Learning
- Explore important concepts
- Installing Software and Setting Up6:04
This is the first step with few practical tasks to get started. Yes! We are talking about installation and initial set up, which we’ll be doing together in this video.
- Install Python
- Set your system
- Understanding NLP4:10
Natural language processing or NLP, is a significant subfield of machine learning, which deals with the interactions between machine and human natural languages. We will explore this, through this video.
- Explore about NLP and its real world applications
- Touring Powerful NLP Libraries in Python10:36
After a short list of real-world applications of NLP, we will be touring the essential stack of Python NLP libraries in this video.
- Download NLTK packages
- Import built-in stemmer algorithms, and initialize a stemmer
- Getting the Newsgroups Data3:17
Let’s get a step ahead by getting the Newsgroups data and downloading it.
- Import the loader function for the 20 newsgroups data
- Download the dataset with the default parameters
- Thinking about Features5:31
This video will walk you through the steps to achieve the extraction of features from the dataset.
- Use the unique function from NumPy
- Get the distinct values of integers
- Look at the first document, its topic number, and name
- Visualization3:15
Let’s get a general idea of how the data is structured, what possible issues may arise, and if there are any irregularities that we have to take care of.
- Display the distribution of the classes
- Display a histogram of the 500 highest word counts
- Data Preprocessing2:29
This video will let you improve the most indicative Features from visualization by using the data preprocessing Techniques.
- Get the 500 words with highest counts
- Apply filtering
- Clustering4:04
How could you find the best division or a decent Approximation between set of data in a dataset? That’s what we are going to learn in this video.
- Apply clustering to the newsgroup dataset
- Display a scatter plot of the actual labels and the cluster labels
- Topic Modeling3:40
When we read a text, we expect certain words appearing in the title or the body of the text to capture the semantic context of the document. This video, will let you achieve this in Machine Learning and Python programming.
- Apply NMF to document clustering and signal processing
- Getting Started with Classification5:02
It is a great starting point of learning classification with a real-life example-our email service providers are already doing this for us, and so can we. We will be learning the fundamental and important concepts of classification.
- Explore different types of classification
- Learn the applications of text classification
- Exploring Naïve Bayes2:49
How can we make a prediction of probability distribution over all classes, besides the most likely class that the data sample is associated with? Let’s answer this question, through this video!
- Look at examples to understand the naive bayes theorem
- The Mechanics of Naïve Bayes4:49
This video will let you understand the magic behind the algorithm-how naive Bayes works.
- Learn the operation of Naïve Bayes algorithm
- The Naïve Bayes Implementation16:30
The most important task after learning a concept is to implement it. Let’s try the implementation of Naïve Bayes algorithm with some sample codes, in this video.
- Read all of the email text files
- Keep the spam class information in the label variabl
- Use a function to perform text cleaning
- Classifier Performance Evaluation10:22
Beyond accuracy, there are several measurements that give us more insights and avoid class imbalance effects. Let’s see these performance measures right now!
- Use the confusion matrix function
- Model Tuning and cross-validation5:11
Having learned what metrics are used to measure a classification model, we can now study how to measure it properly. Let’s do it right away!
- Use the split method and initialize a 10-fold generator
- Repeat the process of term count feature extraction, classifier training, and prediction
- Recap and Inverse Document Frequency4:26
This video will show you how to adopt a more comprehensive approach to extract text features, the term frequency-inverse document frequency
- Use the tfidf approach to extract text features
- The Mechanics of SVM5:51
Since there can be infinite number of feasible Hyperplanes, how can we identify the optimal one? Let's do it in this video!
- Explore SVM in detail through a different scenarios
- The Implementations of SVM5:44
Let's put into action, the fundamentals of SVM Classifier right away on news topic classification and learn to deal with more than two classes.
- Apply the SVM classifier right away on news topic classification
- The Kernels of SVM3:13
What could you do if you are not able to find any linear hyperplane to separate two classes? Let’s see the solution to this problem!
- Visualize the dataset with corresponding decision boundary
- Choosing Between the Linear and the RBF Kernel3:51
This video will walk you through different scenarios where the linear kernel is favored over RBF.
- Work with three scenarios where the linear kernel is preferred over RBF
- News topic Classification with Support Vector Machine10:26
It is finally time to build our state-of-the-art, SVM-based news topic classifier with all we just learned. Let’s do it!
- Load and clean the news dataset
- Tune the penalty C via cross-validation
- Apply the fit method, and record the running time
- Fetal State Classification with SVM5:59
After a successful application of SVM with the linear kernel, let’s look at one more example where SVM with the RBF kernel is suitable for it
- Read the data located in the sheet, Raw Data
- Check class proportions and tune the RBF-based SVM model
- Employ the optimal model to the testing set and check the performance for individual classes
- Brief Overview of Advertising Click-Through Prediction4:45
How can you find whether a given ad on a given page or app will be clicked by a given user or not, with predictive features? Let’s answer this question.
- Decision Tree Classifier13:43
When working on a prediction model, how can you get all of the possible decision alternatives and the corresponding outcomes? Let’s see this, now.
- Learn the Construction of a decision tree
- Calculate the Gini impurity
- Implement the entropy calculation function
- The Implementations of Decision Tree6:17
With a solid understanding of partitioning evaluation metrics, let's practice the CART tree algorithm by hand on a simulated dataset.
- Calculate the weighted impurity of two potential children
- Define a utility function to split a node into left and right child
- Define the greedy search function
- Click-Through Prediction with Decision Tree6:52
After several examples, it is now time to predict ad click-through with the decision tree algorithm we just thoroughly learned and practiced.
- Use the dataset from a Kaggle machine learning competition
- Read the dataset
- Use the model with the optimal parameter to predict unseen cases
- Random Forest - Feature Bagging of Decision Tree5:13
How will you reduce the high variance that a decision tree model suffers from and hence in general performs better than a single tree? Let’s see the solution for this problem now!
- Tweak the max_depth parameter
- Use the model with the optimal parameter None for max_depth
- One-Hot Encoding - Converting Categorical Features to Numerical6:00
This video will show you how one-hot encoding converts the categorical feature to k binary features
- Transform the original features to the encoded features
- Transform the encoded features back to the original features
- Use LabelEncoder to convert a categorical feature to an integer feature
- Logistic Regression Classifier12:09
Let’s turn to a new algorithm with high scalability to large datasets. Yes! We are talking about Logistic regression which is one of the most scalable classification algorithms.
- Define the logistic function
- Declare the input variables from -8 to 8, and the output correspondingly
- Click-Through Prediction with Logistic Regression by Gradient Descent21:30
Let’s deploy the algorithm that we just developed and test it in our click-through prediction project.
- Train a logistic regression model via stochastic gradient descent
- Train a logistic regression model with regularization
- Train on large-scale datasets with online learning
- Feature Selection via Random Forest4:29
How can we rank the importance of features based on their occurrences in nodes among all trees, and select the top most important ones? Let’s see this right now!
- Fit the random forest model
- Look at the bottom 10 features and the corresponding 10 least important features
- Select the top 500 features
- Brief Overview of the Stock Market And Stock Price3:48
Let’s explore the relationships between the observations and the targets, and to output a continuous value based on the input features of an unknown sample.
- Understand the basics of stock market
- Learn how regression is used to model stock market data
- Predicting Stock Price with Regression Algorithms6:41
This video will walk you through some essential steps to apply regression techniques in predicting prices of a particular stock.
- Explore how to develop price prediction models
- Learn about feature engineering
- Data Acquisition and Feature Generation3:30
Let’s get started with obtaining the dataset we need for our project.
- Implement the function for feature generation
- Linear Regression8:27
Since all features and driving factors are available, let’s focus on regression algorithms that estimate the continuous target variables from these predictive features.
- Define the function computing the prediction
- Continue with the function updating the weight w by one step
- Train a linear regression model by 5000 iterations
- Decision Tree Regression7:36
After linear regression, the next regression algorithm need to learn is decision tree regression. Let’s start exploring it right away!
- Define the MSE and weighted MSE computation functions
- Use the DecisionTreeRegressor package from scikit-learn
- Use the regression forest package, RandomForestRegressor
- Support Vector Regression2:57
Let’s get started with the third regression algorithm which is support vector regression (SVR).
- Solve the house price prediction problem with SVR
- Regression Performance Evaluation3:27
So far, we have covered several popular regression algorithms in-depth and implemented them from scratch by using existing libraries. Let’s step ahead and evaluate the performance of these regression algorithms.
- Fine tune the parameters of the linear regression model
- Obtain the optimal set of parameters and predict the testing set with the optimal model
- Evaluate the performance on testing sets based on metrics MSE, MAE, and R2.
- Stock Price Prediction with Regression Algorithms9:49
Now that we have learned three commonly used and powerful regression algorithms and performance evaluation metrics, why don't we utilize all of these in solving our stock price prediction problem? Let’s do it in this video.
- Select the best regression forest model and make a prediction of the testing samples
- Measure the prediction performance
- Use the rescaled data to train the SVR model
- Best Practices in Data Preparation Stage11:17
Apparently, no machine learning system can be built without data. Data collection should be our first focus. Let’s learn the best practices for the data preparation stage.
- Understand the project goal
- Collect all the fields that are relevant
- Maintain consistency of field values and deal with missing data
- Best Practices in the Training Sets Generation Stage8:45
With well-prepared data, it is safe to move on with the training sets generation stage. Let’s move ahead and explore this.
- Determine categorical features with numerical values
- Decide whether or not to encode categorical features
- Decide whether or not and how to select features
- Best Practices in the Model Training, Evaluation, and Selection Stage3:40
Given a machine learning problem, the first question many people ask is usually: what is the best classification/regression algorithm to solve it? This video will bring to you an answer for this question.
- Choose the right algorithm and reduce overfitting
- Diagnose overfitting and underfitting
- Best Practices in the Deployment and Monitoring Stage5:39
After all the processes in the previous three stages, we now have a well established data preprocessing pipeline and a correctly trained prediction model. The last stage of a machine learning system involves saving those resulting models from previous stages and deploying them on new data, Let’s see how to do this.
- Save, load, and reuse models
- Monitor model performance
- Update models regularly
- The Course Overview3:48
This video gives glimpse of the entire course.
- Defining Machine Learning4:15
This video is an introduction to machine learning.
- Understand what machine learning is
- Look at types of machine learning and its tasks
- Training Data, Testing Data, and Validation Data2:27
In this video, we will learn about the three important datasets in machine learning.
- Understand training, testing, and validation data
- Look at the cross validation method
- Bias and Variance3:25
Many performance metrics can be used to measure the amount of prediction error. Here, we will look at bias and variance.
- Understand the concept of bias and variance
- Learn how to measure accuracy, precision and recall of a model
- An Introduction to Scikit-learn3:30
Scikit-learn has become one of the most popular machine learning libraries. We will learn about it in this video.
- Look at the uses of scikit-learn
- Install scikit using pip and Anaconda
- Installing Pandas, Pillow, NLTK, and Matplotlib2:46
In this video, we will install the supporting libraries for scikit.
- Understand the importance of these libraries
- Install pandas, NLTK, Pillow, and matplotlib using a single command
- What Is Simple Linear Regression?8:00
Simple linear regression can be used to model the linear relationship between one response variable and one feature. Let’s see how to do that.
- Visualize training data and enter the pizza price predictor algorithm using simple linear regression
- Evaluate the fitness of the model
- Solve OLS for simple linear regression
- Evaluating the Model2:22
We have estimated a model’s parameters. But how do we know if our model is a good representation of the real relationship? We will evaluate the model in this video.
- Find total sum of squares
- Find RSS
- Calculate R-squared
- KNN, Lazy Learning, and Non-Parametric Models4:08
KNN is a simple model for regression and classification tasks. Also, it is a lazy learning model and non-parametric model. Let’s learn more about it in this video.
- Understand how KNN models work
- Understand the concept of lazy learning and non-parametric models
- Classification with KNN7:38
In this video, we will learn classification with KNN. We will solve a toy classification problem where we have to predict a person’s sex based on his or her height.
- Create a scatterplot with matplotlib
- Identify the most common sex of the nearest neighbors
- Implement a KNN classifier
- Regression with KNN4:12
Following the previous video, we will go one step further and use KNN for regression tasks. Here, we will predict a person’s weight based on their height and sex.
- Calculate MAE and MSE
- Scale features and standardize them using StandardScaler transformer
- Fit the regressor and compare performance of two models
- Extracting Features from Categorical Variables1:44
Many examples have explanatory variables that are categorical or nominal. Categorical variable can take one of a fixed set of values. Let’s see how to extract features from them in this video.
- One-hot encode categorical variable
- Assign values to variables
- Standardizing Features1:47
Many machine learning algorithms perform better when they are trained on standardized data. So we will standardize feature vectors in this video.
- Standardize dataset using the scale function
- Extracting Features from Text15:41
Many machine learning problems use text, which usually represents natural language. Text must be transformed to a vector representation that encodes some aspect of its meaning. So that will be done in this video.
- Learn about the bag-of-words model
- Stop word filtering. Perform stemming and lemmatization
- Perform the hashing trick and look at word embedding
- Multiple Linear Regression4:31
When we want to train a model on more attributes, we use multiple linear regression.
- Look at the multiple linear regression model
- Solve for ß using NumPy
- Polynomial Regression2:24
In the previous video, we tried linear regression. Here, we will use polynomial regression a special case of multiple linear regression
- Fit a model to polynomial features
- Compare it to the simple linear regression model
- Plot the quadratic, cubic, and higher order polynomial
- Regularization2:15
Regularization is a collection of techniques that can be used to prevent overfitting. Regularization adds information, often in the form of a penalty against complexity, to a problem. So let’s learn how to perform regularization.
- Look at ridge regression
- Understand LASSO and elastic net regularization
- Applying Linear Regression6:07
We have worked through a toy problem to learn how linear regression models relationships between explanatory and response variables. Now we'll use a real dataset and apply linear regression to an important task.
- Explore the data
- Fit and evaluate the model
- Gradient Descent2:05
In this video, we will discuss another method for efficiently estimating the optimal values of the model's parameters called gradient descent.
- Minimize the cost function using gradient descent
- Predict the prices of houses using SGD algorithm
- Fit and evaluate the estimator
- Binary Classification with Logistic Regression2:47
Ordinary linear regression assumes that the response variable is normally distributed. In logistic regression, the response variable describes the probability that the outcome is the positive case. Let’s work with it.
- Understand logistic regression and look at the logistic function
- Spam Filtering7:35
In modern days, spam has become a reason for concern and a waste of time. We therefore will filter spam in this video.
- Explore the dataset and calculate basic summary statistics
- Make predictions
- Calculate the parameters
- Tuning Models with Grid Search2:04
Grid search is a common method for selecting the hyperparameter values that produce the best model. It is an exhaustive search that trains and evaluates a model for each possible combination of the specified hyperparameter values. So let’s learn how to work with it.
- Use GridSearchCV to find better hyperparameter values.
- Multi-Class Classification4:15
In many classification problems, there might be more than two classes. In such cases, we need multi-class classification.
- Explore the dataset
- Examine the target classes
- Train the classifier and evaluate classifier’s predictions
- Multi-Label Classification and Problem Transformation4:34
The final type of classification problem that we will discuss is multi-label classification, in which each instance can be assigned a subset of the set of classes.
- Understand the concept of problem transformation
- Calculate performance metrics of multi-label classification
- Bayes' Theorem3:33
Bayes' theorem is a formula for calculating the probability of an event using prior knowledge of related conditions. Let’s know more about it here.
- Understand Bayes theorem
- Solve conditional probability
- Generative and Discriminative Models2:05
In this video, we are going to learn about generative and discriminative models.
- Understand the working of discriminative and generative models with examples
- Naive Bayes with Scikit-learn4:03
We have learned about the Naïve Bayes Classifier. We will now fit it with scikit-learn.
- Load the dataset. Split data into training and testing set
- Evaluate the model and plot the scores
- Compare logistic regression with Naïve Bayes classifier
- Decision Trees1:54
Decision trees are tree-like graphs that model a decision. They help you to decide on something using a step-by-step approach.
- Understand how decision trees work
- Training Decision Trees10:08
After learning about decision trees, let’s see how it is trained in this video.
- Select questions
- Calculate information gain and Gini impurity
- Decision Trees with Scikit-learn6:03
Finally, let’s write code for implementing the decision tree in this video.
- Perform grid-search to find hyperparameter values that produce the decision tree
- Evaluate the tree’s performance on a test set
- Look at advantage and disadvantages of the decision trees
- Bagging3:53
Bootstrap aggregating, or bagging, is an ensemble meta-algorithm that can reduce the variance in an estimator. Bagging can be used in classification and regression tasks. Let’s get to know it a little better.
- Create a random forest
- Train a random forest using scikit-learn
- Boosting2:55
Boosting is a family of ensemble methods that are primarily used to reduce the bias of an estimator. It can be used in classification and regression tasks. Let’s see how to work with it.
- Train an AdaBoostClassifier
- Plot the accuracy of the ensemble
- Compare the accuracy with that of a single decision tree
- Stacking2:32
Stacking is an approach to creating ensembles that we talked about in the previous videos. Let’s see how that works.
- Create an artificial classification dataset
- Train and evaluate the individual base estimators
- Train and evaluate the ensemble
- The Perceptron–Basics6:50
The perceptron is the most basic unit of the neural network. We need to know about it to work on neural networks further.
- Learn about types of activation functions and the perceptron leaning algorithm
- Perform document classification with neural networks
- Limitations of the Perceptron2:12
After learning to work with a perceptron, we should be aware of its limitations too.
- Look at the limitations of perceptron with an example
- Kernels and the Kernel Trick4:38
Sometimes we do not require the mapped feature vectors after scalar is computed. We can compute scalar without using the dot product of mapped vectors using the kernel trick. A kernel is a function that, given the original feature vectors, returns the same value as the dot product of its corresponding mapped feature vectors.
- Look at the kernel formula
- Work with an example with values for feature vectors
- Maximum Margin Classification and Support Vectors3:36
In classification, in order to know which classification algorithm works better, we need to know which boundary is the best. For that, we will learn about margin classification in this video.
- Calculate the functional margin
- Maximize the geometric margin
- Classifying Characters in Scikit-learn6:44
In this video, we will use SVMs to classify characters in scikit-learn.
- Classify handwritten digits
- Classify characters in natural images
- Nonlinear Decision Boundaries2:09
Sometimes some functions are not linear and the data cannot be classified with linear classification. We require nonlinear decision boundaries for that.
- Look at an example to understand nonlinear decision boundaries
- Feed-Forward and Feedback ANNs1:35
There are two main types of ANN, feedback and Feed-forward. They are used depending on the application and need. We will learn about them in this video.
- Learn about the two main artificial neuron types
- Multi-Layer Perceptrons and Training Them8:30
To increase the efficiency of the neuron, multiple layers are added to it. They are called multi-layer perceptrons and this video will all about working with them
- Understand the multi-layer perceptron
- Train the MLP to approximate XOR and classify handwritten digits
- Clustering1:33
Clustering, or cluster analysis, is the task of grouping observations so that members of the same group, or cluster, are more similar to each other by some metric than they are to members of other clusters.
- Understand the concept of clustering with a dataset
- K-means5:29
The K-means algorithm is a clustering method that is popular because of its speed and scalability. Let’s get started with it.
- Calculate local optima
- Select k with the elbow method
- Evaluating Clusters1:47
We can perform evaluation of unsupervised algorithms using intrinsic measures. We will introduce the silhouette co-efficient in this video, which is a performance measure.
- Run code for forming clusters of different sizes
- Calculate the silhouette co-efficient for each
- Image Quantization2:27
Image quantization is a lossy compression method that replaces a range of similar colors in an image with a single color. Quantization reduces the size of the image file since fewer bits are required to represent the colors. Let’s do it!
- Read and flatten the image
- Use K-means to create clusters. Predict the cluster assignment
- Create the compressed image
- Principal Component Analysis6:38
Learning from large data required more memory and processing power. PCA is a technique used for finding patterns in high dimensional data. So it saves the extra effort.
- Learn about variance covariance and covariance metrics
- Study eigenvectors and eigenvalues
- Understand how to perform PCA
- Visualizing High-Dimensional Data and Face Recognition with PCA5:29
A high dimensional dataset cannot be represented graphically, but we can still gain some insights into its structure by reducing it to two or three principal components. Let’s do that using PCA.
- Load the iris dataset and instantiate a PCA estimator
- Reduce the data matrix
- Assemble and plot reduced data
- The Course Overview3:48
This video will provide an overview of the entire course.
- Introducing Deep Learning3:59
Get the user excited about ML; a quick view of various platforms and key products using the ML tools (photo search, Siri, and so on).
- Introduce TensorFlow
- Installing TensorFlow on Mac OSX3:50
We will install the TensorFlow platform and associated prerequisites on the Mac OS X operating system.
- Install the Python package manager PIP
- Install dependencies such as SIX
- Install TensorFlow
- Installation on Windows – Pre-Reqeusite Virtual Machine Setup2:49
Since TensorFlow does not function natively on Windows, we will cover Virtual Machine setup so Windows users can run Linux on a VM.
- Set up a virtual machine
- Allocate the right amount of space for our installation and projects
- Obtain and install an appropriate Linux installation image
- Installation on Windows/Linux4:00
We will cover TensorFlow installation on Linux. This also covers installation on Windows via a Virtual Machine running Linux.
- Set up a virtual machine
- Obtain an appropriate Linux installation image
- The Hand-Written Letters Dataset3:01
Which dataset do we use for our first classifier, and what attributes do we consider? Let's use a dataset of letters in numerous typefaces.
- Find a suitable dataset for training our classifier
- Review dataset attributes
- Compare between MNIST and notMNIST datasets
- Automating Data Preparation3:20
We just spent a good amount of time obtaining and preparing data, are we expected to do this for each run? No, you'll now learn to automate the initial steps used to feed the machine learning process.
- Download data automatically
- Extract data automatically
- Load training and test images into a processable format
- Understanding Matrix Conversions
How do we represent images internally within the machine learning environment? We'll hold image data in a stack of matrices.
- Load images into memory
- Shuffle the images
- Segment out distinct working sets from our images
- The Machine Learning Life Cycle1:51
So now that we have looked upon the major concepts, the question remains how do we actually train the classifier in code? We are now going to code and train a classifier using TensorFlow.
- Understand the life cycle
- Design the neural network
- Train the system
- Reviewing Outputs and Results2:51
Once training is complete, how are the results interpreted and measured? We'll use the training set, with known labels to measure how well our trained system performs.
- Use a trained network to predict classes for the test set
- Use validation methods to compare predictions against actuals
- Find the accuracy of our classifier on the test set
- Getting Started with TensorBoard5:08
How do we monitor the internals of our training setup and execution? We'll use TensorBoard to view our network architecture and probe values through the training.
- Get TensorBoard up and running
- Explorer Events and Histograms functionality
- Learn Graph Explorer
- TensorBoard Events and Histograms5:21
How do we monitor the scalars and tensors on TensorBoard? We'll set up code for log specific variables and TensorBoard reads these structured logs.
- Use specific commands to capture interesting variables in a structured manner
- Point TensorBoard toward these variables
- Browse through, view, and work with individual variable graphs
- The Graph Explorer5:08
How do we view the network architecture we've built for our training? We'll push the graph to a TensorBoard readable log and view it on Graph Explorer.
- Use specific code, push the graph to TensorBoard logs
- View the network architecture on Graph Explorer
- Zoom into individual components, view parameters, inputs, and outputs
- Our Previous Project on TensorBoard5:02
How do we actually apply what you've just learned to our own project? You'll learn to go through individual changes to our previous project.
- Initialize TensorBoard variables
- Capture the graph as well as interesting variables
- Execute training and launch TensorBoard to view outputs
- Fully Connected Neural Networks4:44
How do we process images with more complex and better suited neural networks? For this purpose, we will use the well-known convolutional neural networks also known as CNNs.
- Learn the difference between fully connected networks and CNNs
- Learn how CNNs work better than normal NNs
- Change our previous code to add CNN model for training
- Convolutional Neural Networks4:58
How do CNNs work and what does their model look like? We are going to talk about the basic model architecture of CNN.
- Talk about what convolution is
- Discuss what kernels are
- Discuss CNNs architecture
- Programming a CNN5:01
How to implement a basic CNN? We will look into the coding of a CNN.
- Implement the model
- Add appropriate layers
- Talk about the code architecture in TensorFlow
- Using TensorBoard on Our CNN1:58
How does it look after coding? We will visualize the implemented CNN in TensorFlow.
- Add TensorBoard code at appropriate places
- Run the graph with TensorBoard
- Visualize it after the run
- CNN Versus Fully Connected Network Performance2:08
How much better CNNs are as compared to normal fully connected networks? We will compare the performances of the two models on CIFAR-10 dataset.
- Find test accuracy using FCN
- Find accuracy using CNN
- Compare the performance differences
Requirements
- Familiarity with Machine Learning fundamentals will be useful.
- A basic understanding Python programming is assumed.
Description
Machine learning brings together computer science and statistics to build smart, efficient models. Using powerful techniques offered by machine learning, you’ll tackle data-driven problems. The effective blend of Machine Learning with Python, scikit-learn, and TensorFlow, helps in implementing solutions to real-world problems as well as automating analytical model.
This comprehensive 3-in-1 course is your one-stop solution in mastering machine learning algorithms and their implementation. Learn the fundamentals of machine learning and build your own intelligent applications. Explore popular machine learning models including k-nearest neighbors, random forests, logistic regression, k-means, naive Bayes, and artificial neural networks
Contents and Overview
This training program includes 3 complete courses, carefully chosen to give you the most comprehensive training possible.
This course will help you discover the magical black box that is Machine Learning by teaching a practical approach to modeling using Python, scikit-learn and TensorFlow.
The first course, Step-by-Step Machine Learning with Python, covers easy-to-follow examples that get you up and running with machine learning. In this course, you’ll learn all the important concepts such as exploratory data analysis, data preprocessing, feature extraction, data visualization and clustering, classification, regression, and model performance evaluation. You’ll build your own models from scratch.
The second course, Machine Learning with Scikit-learn, covers effective learning algorithms to real-world problems using scikit-learn. You’ll build systems that classify documents, recognize images, detect ads, and more. You’ll learn to use scikit-learn’s API to extract features from categorical variables, text and images; evaluate model performance; and develop an intuition for how to improve your model’s performance.
The third course, Machine Learning with TensorFlow, covers hands-on examples with machine learning using Python. You’ll cover the unique features of the library such as data flow Graphs, training, and visualization of performance with TensorBoard—all within an example-rich context using problems from multiple sources.. The focus is on introducing new concepts through problems that are coded and solved over the course of each section.
By the end of this training program you’ll be able to tackle data-driven problems and implement your solutions as well as build efficient models with the powerful yet simple features of Python, scikit-learn and TensorFlow.
About the Authors- Yuxi (Hayden) Liu is currently an applied research scientist focused on developing machine learning models and systems for given learning tasks. He has worked for a few years as a data scientist, and applied his machine learning expertise in computational advertising. He earned his degree from the University of Toronto, and published five first-authored IEEE transaction and conference papers during his research. His first book, titled Python Machine Learning By Example, was ranked the #1 bestseller in Amazon India in 2017. He is also a machine learning education enthusiast.
- Shams Ul Azeem is an undergraduate in electrical engineering from NUST Islamabad, Pakistan. He has a great interest in the computer science field, and he started his journey with Android development. Now, he’s pursuing his career in Machine Learning, particularly in deep learning, by doing medical-related freelancing projects with different companies. He was also a member of the RISE lab, NUST, and he has a publication credit at the IEEE International Conference, ROBIO as a co-author of Designing of motions for humanoid goalkeeper robots.
Who this course is for:
- Anyone interested in entering the data science stream with Machine Learning.
- Software engineers who want to understand how common Machine Learning algorithms work.
- Data scientists and researchers who want to learn about the scikit-learn API.