Intro to pre-trained ML APIs in Google Cloud Platform

Let's try to understand Google's machine learning APIs under 5 minutes

Chapters:
0:00 Intro
0:25 Machine Learning is hard
1:55 Introducing Google Pre-Trained ML APIs
2:12 Four Main categories of Pre-Trained ML APIs
2:46 Vision API Example
4:16 Outro

Machine Learning is COMPLEX!
Even though ML tops the list of being a buzzword, in almost every sales pitch

So, There goes a lot into training of these Neural Network to solve relatively basic problems.Like distinguishing dogs and cats

To power any simple model like this a neural network with multiple layers has to be trained on series dataset for it to arrive to being usable.

While training, each image is broken into pixel that is fed to the NN to create layers of weights and biases, that can later predict in the future

Furthermore, you need 100s or thousands of images to train a relatively usable model

API stands for Application Programming Interface. An API is a software intermediary that allows two applications to talk to each other. In other words, an API is the messenger that delivers your request to the provider that you’re requesting it from and then delivers the response back to you.
These are a set of pre-trained APIs to give you access to pre-trained machine learning models with a single REST API request.

Thankfully google cloud platform offers several high quality pre-trained models, via APIs that can be easily be consumed by developers regardless of the platform

The ML APIs available can be divided into 4 main categories, +1 more category as they are coming out with new ones

Four main categories are
Sight
Language
Conversation
Structured Data

LINKS:

FOLLOW ME ON

#gcp #machinelearning #googleapis #googelcoudplatform #ai #ml #vision #ml

[sunnylew]

Clear explanation! Love it!
Hope that will have more videos regarding ML APIs!
Keep Going!