What are the five steps to convert a potential case into a resource that can be exploited using machine learning technology?
Why is it important not to skip steps?
Why are neural networks so popular?
How can one present a supervised learning problem and find a good general solution using gradient descent and a measured way of creating datasets?
Learn to write machine learning models
Distributed models that will scale in TensorFlow and provide high-quality predictions.
Transform raw data into features in a way that allows the machine to learn important characteristics of the data and bring human insights for problem solving.
Learn to combine the appropriate parameters that will lead to the development of accurate and generalizable models.
Get to know the theory for solving different types of problems in AA.
End-to-end machine learning experience
Starting with building a strategy focused on AA.
Progress towards training, optimization, and model generation.
Hands-on labs using Google Cloud Platform.
By joining this internship
You accept the Qwiklabs Terms of Service as stated in the FAQ.
This internship includes hands-on labs using our Qwiklabs platform.
The practical components will allow you to apply the skills you gain in the video lessons.
Projects include topics such as Google Cloud Platform products used and configured in Qwiklabs.
You will gain practical experience with the terms explained in all modules.
Details of the courses that make up the specialization
How Google is doing machine learning in Spanish courses
Course 1
15 hours
4.6 (239 ratings)
Course Details
What you will learn:
Describe the Vertex AI platform and how it is used to create, train, and deploy AutoML learning models without writing a line of code.
Describe the recommendations for implementing machine learning in Google Cloud.
Take advantage of Google Cloud Platform’s tools and environment to perform AA.
Define the recommendations for responsible IA.
Course 2
13 hours
4.7 (127 ratings)
Course Details
What you will learn:
Describe how to improve data quality and perform exploratory data analyses.
Collect and train AutoML models with Vertex AI and BigQuery.
Optimize and evaluate the models using loss functions and performance criteria.
Create training, evaluation, and testing datasets that can be repeated and expanded.
Course 3
18 hours
4.5 (134 ratings)
Course Details
What you will learn:
The course focuses on the flexible and easy use of TensorFlow 2.x and Keras for creating, training, and deploying machine learning models.
You will learn about the structure of the TensorFlow 2.x API and become familiar with the key components of TensorFlow through hands-on exercises.
We will teach you how to work with data sets and attribute columns.
You will learn to design and build a data pipeline in TensorFlow 2.x.
You will gain hands-on experience loading NumPy arrays, images, and text data with tf.data.Dataset, as well as CSV data with Pandas.
You will also gain practical experience in creating numeric attribute columns, categories, and definitions with clipping.
We will also introduce you to Keras’ sequential and functional API for building deep learning models.
We will talk about activation, loss, and optimization functions.
Our hands-on labs on Jupyter notebooks will allow you to build machine learning models of basic linear regression and basic and advanced logistic regression.
You will learn to train, run, and deploy machine learning models at scale with Cloud’s AI Platform.
What you will learn to acquire:
Category: TensorFlow
Category: Python Programming
Category: Machine Learning
Category: Feature Engineering
Course 4
9 hours
4.3 (35 ratings)
Course Details
What you will learn:
Describe the Vertex AI matching store and compare the key aspects that make the feature useful.
Perform feature engineering with BigQuery ML, Keras, and TensorFlow.
Analyze how to process and explore features with Dataflow and Dataprep.
Course 5
19 hours
4.7 (50 ratings)
Course Details
What you will learn:
AA model rules using regularization techniques.
Adjust the retention view size and learning rate to improve the model’s performance.
