Wednesday, April 14, 2021

How to Train a Deep Learning TensorFlow Analytic to Play Connect 4

Bot Libre now allows you to create generic deep learning analytics and train them through our web API. Deep learning analytics can be used for a wide array of purposes to analyze and make predications on data. This example shows how to train a deep learning analytic to play Connect 4.

You can use either the Bot Libre deep learning library, or the TensorFlow deep learning library. You can choose the inputs, outputs, and layers.

The web API endpoint to train the network is:
https://www.botlibre.com/rest/api/train-data-analytic

The web API endpoint to test the network is:
https://www.botlibre.com/rest/api/test-data-analytic

Create the Analytic

1. Go to the Deep Learning page and create a new analytic.

2. Configure the network with the following settings:

3. Click on "Save", and then "Reset Network".

Train the Analytic

You can train the analytic using the Bot Libre web API

https://www.botlibre.com/rest/api/train-data-analytic

Create a Connect 4 Chatbot

Now that you have an analytic trained to play Connect 4, you can create your own Connect 4 chatbot with your own analytic!

Follow this tutorial to create your Connect 4 chatbot.

Then, modify the connect4 script with your application ID and analytic ID to connect it to your analytic.

Here is an example Conect 4 chatbot for you to try: Games Bot.

How to Create a Chatbot that can Play Connect 4

Creating your own bot that can play Connect 4 can be done in a few simple steps:

1. First you'll need to sign into a Bot Libre account.

2. Once you're signed in, create a bot or go to a bot that you have already made.

3. Go to your bot's Admin Console, then to its Scripts page click on the Import button or menu, and import the script 'Connect 4'.

4. (Optional) By default, the connect4 script will play random moves against you. If you want the connect4 script to use our trained deep learning analytic, you need to make the following changes to the script:

a. Lines 88-89: Comment out randomMove(), and uncomment deepLearningMove().

b. Line 243: Replace the placeholder application ID with your application ID.

c. Line 271: Verify that the POST call is made to the correct URL.

5. By typing "connect 4" in the chat, your bot will be able to create a board to play against you in connect 4.

6. If you want to end the game type, "end".

You can also create a new bot using the games_template or chat with the Games Bot

Monday, April 12, 2021

Detecting a person's age in an image using deep learning and the Bot Libre analytics API

Bot Libre Analytic lets you easily train your own deep learning neural network without any programming. With Bot Libre you can create your own deep learning neural network for image recognition, audio and speech recognition, object detection, games, prediction, data analysis, and more. To create your own analytic see How to create your own deep learning neural network for image recognition

Bot Libre also has many pretrained analytics trained by us and our users that you can use through the Bot Libre web API from your own web or mobile application. To browse Bot Libre's analytics go here.

Here is the Age Detector.


You can access it from the Bot Libre website, or you can access it analytic through the Bot Libre web API, or using the Bot Libre JavaScript SDK.

Here is an HTML example of using the Bot Libre JavaScript SDK:

Image Classifier Website.

 

To use the web API call the end point using a POST HTTP request,

https://www.botlibre.com/rest/api/test-analytic

Pass an XML entity for the analytic,

<analytic application='12345' id='35935580'/>

and a multipart form data for the image.

See https://www.botlibre.com/api.jsp

 

If you would like help using this analytic, or developing your own deep learning solution for your own application contact sales@botlibre.biz

Friday, April 2, 2021

Classifying an image using deep learning, Inception v3, and the Bot Libre analytics API

Bot Libre Analytic lets you easily train your own deep learning neural network without any programming. With Bot Libre you can create your own deep learning neural network for image recognition, audio and speech recognition, object detection, games, prediction, data analysis, and more. To create your own analytic see How to create your own deep learning neural network for image recognition

Bot Libre also has many pretrained analytics trained by us and our users that you can use through the Bot Libre web API from your own web or mobile application. To browse Bot Libre's analytics go here.

Here is the Inception v3 Classifier.

Inception v3 is a convolutional neural network designed for image classification from GoogleNet.


You can access it from the Bot Libre website, or you can access it analytic through the Bot Libre web API, or using the Bot Libre JavaScript SDK.

Here is an HTML example of using the Bot Libre JavaScript SDK:

Image Classifier Website.

 

To use the web API call the end point using a POST HTTP request,

https://www.botlibre.com/rest/api/test-analytic

Pass an XML entity for the analytic,

<analytic application='12345' id='21454183'/>

and a multipart form data for the image.

See https://www.botlibre.com/api.jsp

 

If you would like help using this analytic, or developing your own deep learning solution for your own application contact sales@botlibre.biz

Thursday, April 1, 2021

How to Create a Electronics Store Chatbot

To begin creating your own bot for an electronics store you must first create an account at botlibre.com or botlibre.biz

Once your account has been created, click on "Bots." Here you will be presented with several options, and you will then click on "New Bot" and will be presented with the following page:

In the form, enter a "bot name" and set template to "electronics_store_template" If you have a website you may enter your website URL into the "Website" field. You may leave the other fields empty and click the "Create" button at the bottom.

Your bot is now created. To give your bot set responses first click "My Bots" from the "Bots" page, and then click on your bot. Next click on "Admin Console" (the gear icon).

By clicking on "Embed" you can generate a script which can be embedded on a site like Blogger in an HTML/JavaScript gadget.

Once in "Admin Console" click on "Training and Chat Logs"

In Training and Chat Logs you can now:

       a) Add, review responses

       b) Add, review greetings

       c) Add, review default responses & more

Here is the electronics store bot's response list.

To test it out try our demo bot: Electronics Store Bot

To see this bot on a website template, check out this link: Electronics Store

If you have any issues with setting up your bot you may contact us at our email at support@botlibre.com or upgrade to our Platinum Service and we can create your bot for you.