Salesforce Einstein Vision: How it works?

Salesforce Einstein Vision: How it works?

Before starting about this, Please give few minutes to this article What is AI and Machine Learning?

If you have done with above blog. Let’s cover Salesforce Einstein Vision.

Salesforce Einstein Vision: Salesforce Einstein Vision is a powerful APIs with great machine learning algos. Salesforce has two versions of these APIs. V2 is in beta. Here are they1. Salesforce Einstein Version-1 (v1) comes with Image classification APIs2. Salesforce Einstein Version-2 (v2) comes with Natural Language Processing (Get the sentiment of the sentence and Intent processing). + V1

Salesforce Einstein Vision is all about following five terms:1. Dataset2. Labels3. Training4. Models5. Prediction

Let’s take an example of Image Classifier with Salesforce Einstein Vision. We are going to predict about fruits. So let’s assume we have 3 fruits here with their few images (Mango, Banana, Grapes).

1. Dataset and Labels:

Here Fruit is the dataset and labels are the names of the fruits.


Note: Related APIs->,

2. Creating and Training Model:

Now next process is Train the data and creates the model from our dataset. We will pass this dataset to Train API. Here we can decide how much data we need to use as Training data and Testing data. Train API have trainParams={“trainSplitRatio”: 0.n} flag in the API. Here n is Trading data. If you write 0.7 that means, 70% data will be used as training data and 30% will be used as test data.


Note: If you are not clear about Training and Testing data. Please read this What is AI and Machine Learning?? blog.

Note: Related APIs->

3. Prediction:

Now it’s time to rock with Salesforce Einstein. We will send an image of banana to Salesforce Einstein with our model of fruits. It will return the predicted output as Banana.


Note: Related APIs->

I hope this will give you some basic startup with Salesforce Einstein Vision.


Popular Salesforce Blogs