Machine Learning

Course Details
Course Syllabus
Expert Overview
Schedule A Demo

About ML

Are you a fresher about to dive into the world of virtual wonder? Are you a professional raising yourself to reach your ambition? You have arrived at your destination. As the name gives, join us to know what makes the machine learn and see how computers attain the ability to learn. ✨

Get the best out of hands-on lab experience with our experts – create your mini project – in Customer Relation Management, Business Intelligence, and much more!!

Become an important asset in the corporate world, by giving companies trends and operational patterns in data. Develop and deliver new thrills that leave an impact.

 Why learn ML anyway?

Gone are the days when there was a lack of information. Data is now the pulse of all business. The difference between going forward or being left behind in a cutthroat market is information. And by leveraging the amazingly versatile ML you can unlock the key to being ahead.


Upskill yourself to get into major industries- and as reported by LinkedIn ML engineering is the fourth fastest-growing job over the past five years. So take advantage of the growing market! And upskill yourself today.

Why is ML worth it?

As the world progresses, convenience is the new norm. To drive a car? We use ML (Tesla). To watch a movie? ML (Netflix). Turn off the lights? ML (Google Home). It is all around us! And it is inescapable.

And as this grows, so does the industry surrounding it. By 2025 the industry is expected to grow to $19.9 billion. Don’t miss the opportunity to be the brain behind business intelligence.


Want to take yourself a step further? Check out our AI courses!

Domain 1:

Data Engineering

 

1.1  Create data repositories for machine learning.

·         Identify data sources (e.g., content and location, primary sources such as user data)

·         Determine storage mediums (e.g., DB, Data Lake, S3, EFS, EBS)

1.2  Identify and implement a data ingestion solution.

· Data job styles/types (batch load, streaming)

· Data ingestion pipelines (Batch-based ML workloads and streaming-based ML workloads)

o Kinesis

o Kinesis Analytics

o Kinesis Firehose

o EMR

o Glue

· Job scheduling

1.3 Identify and implement a data transformation solution.

 · Transforming data transit (ETL: Glue, EMR, AWS Batch)

 · Handle ML-specific data using map reduce (Hadoop, Spark, Hive)

 

Domain 2:

Exploratory Data Analysis

 

2.1 Sanitize and prepare data for modelling.

·         Identify and handle missing data, corrupt data, stop words, etc.

·         Formatting, normalizing, augmenting, and scaling data

·         Labelled data (recognizing when you have enough labelled data and identifying mitigation strategies [Data labelling tools (Mechanical Turk, manual labour)])

2.2 Perform feature engineering.

·         Identify and extract features from data sets, including from data sources such as text, speech, image, public datasets, etc.

·         Analyse/evaluate feature engineering concepts (binning, tokenization, outliers, synthetic features, 1 hot encoding, reducing dimensionality of data)

2.3 Analyse and visualize data for machine learning.

·         Graphing (scatter plot, time series, histogram, box plot)

·         Interpreting descriptive statistics (correlation, summary statistics, p value)

·         Clustering (hierarchical, diagnosing, elbow plot, cluster size)

 

Domain 3:

Modelling

3.1 Frame business problems as machine learning problems.

·         Determine when to use/when not to use ML

·         Know the difference between supervised and unsupervised learning

·         Selecting from among classification, regression, forecasting, clustering, recommendation, etc.

 3.2 Select the appropriate model(s) for a given machine learning problem.

·         Xgboost, logistic regression, K-means, linear regression, decision trees, random forests, RNN, CNN, Ensemble, Transfer learning

·         Express intuition behind models

3.3 Train machine learning models.

·         Train validation test split, cross-validation

·         Optimizer, gradient descent, loss functions, local minima, convergence, batches, probability, etc.

·         Compute choice (GPU vs. CPU, distributed vs. non-distributed, platform [Spark vs. non-Spark])

·         Model updates and retraining o Batch vs. real-time/online

3.4 Perform hyper-parameter optimization.

·         Regularization o Drop out

o   L1/L2

·         Cross validation

·         Model initialization

·         Neural network architecture (layers/nodes), learning rate, activation functions

·         Tree-based models (# of trees, # of levels)

·         Linear models (learning rate) 3.5 Evaluate machine learning models.

·         Avoid overfitting/under-fitting (detect and handle bias and variance)

·         Metrics (AUC-ROC, accuracy, precision, recall, RMSE, F1 score)

·         Confusion matrix

·         Offline and online model evaluation, A/B testing

·         Compare models using metrics (time to train a model, quality of model, engineering costs)

·          Cross validation

 

 

Domain 4:

Machine Learning Implementation and Operations

4.1 Build machine learning solutions for performance, availability, scalability, resiliency, and fault tolerance.

·         AWS environment logging and monitoring

o   CloudTrail and CloudWatch

o   Build error monitoring

·         Multiple regions, Multiple AZs

·         AMI/golden image

·         Docker containers

·         Auto Scaling groups

·         Rightsizing

o   Instances

o   Provisioned IOPS

o   Volumes

·         Load balancing

·         AWS best practices

4.2 Recommend and implement the appropriate machine learning services and features for a given problem.

·         ML on AWS (application services) o Poly o Lex o Transcribe

·         AWS service limits

·         Build your own model vs. SageMaker built-in algorithms

·         Infrastructure: (spot, instance types), cost considerations

o   Using spot instances to train deep learning models using AWS Batch

 

4.3 Apply basic AWS security practices to machine learning solutions.

·         IAM

·         S3 bucket policies

·         Security groups

·         VPC

·         Encryption/anonymization

4.4 Deploy and operationalize machine learning solutions.

·         Exposing endpoints and interacting with them

·         ML model versioning

·         A/B testing

·         Retrain pipelines

·         ML debugging/troubleshooting

o   Detect and mitigate drop in performance

o   Monitor performance of the model