India DreamIn17 Presentation/Code – Salesforce Einstein Vision Implementation

It was really nice to be part of the IndiaDreamIn17 as the Speaker. I presented Salesforce Einstein Vision Implementation. The presentation is explaining about following things: 1. AI & Machine Learning - Read here 2. Einstein Vision - Read Here 3. Dataset, Label, Examples - Read here 4. Training, Model, Prediction - Read here 5. Einstein Key… Continue reading India DreamIn17 Presentation/Code – Salesforce Einstein Vision Implementation


AI Buddy – Release v1.5

Download v1.5 from here. Features:  1. Added View Global Datasets checkbox on Dataset list page. 2. Added new epochs and learning rate parameter for training model. Bug fixes: 1. The app was not running after key upload for new einstein platform users. 2. Dataset upload failing in v1 for image type. Updated document for this release:… Continue reading AI Buddy – Release v1.5

What is Training in Machine Learning?

Training a process where we create a model using our datasets. Every prediction query goes to Model, not to the dataset. Let's assume we have 100 records. We are giving 70% of data to our machine as the Training data and rest of the 30% data as the test data. Got confused what is train… Continue reading What is Training in Machine Learning?

Salesforce Einstein Introduction

The Einstein API allows you to go into the power of AI and train deep learning models to recognize and classify images and sentiments of text. You can use pre-trained classifiers or train your own custom classifiers according to our requirement. For example, A government office asks users to upload the document like passport, unique… Continue reading Salesforce Einstein Introduction

Testing – Test the Einstein Model

  On this page, you can test your model. Copy model ID from your model and paste it in Model ID on this page. Your model should be successfully trained. Otherwise, you will not see prediction. Custom Models: Vision: You must give Model ID of the image model. You can do a prediction for URL image, Base… Continue reading Testing – Test the Einstein Model

Training – Create Model

Go to Dataset Detail page. Here click on Create Model button. It will open a new page asking for Model Name. Fill the following: 1. Model Name: Name of the model. Maximum length is 180 characters. 2. Epochs: Optional. The number of training iterations for the neural network. Valid values are 1–100 (For V2 it is… Continue reading Training – Create Model