Visual Recognition with Watson Studio
  • Building Custom Visual Recognition Models with Watson Studio
  • Getting Started
  • Workshop
    • Setup
    • Train
    • Test
  • Next Steps
Powered by GitBook
On this page
  • Step 1:
  • Step 2 :
  • Step 3:
  • Step 4 :

Was this helpful?

  1. Workshop

Train

This example we are going to create an Hand-Bag Classifier Custom Visual Recognition Model

PreviousSetupNextTest

Last updated 5 years ago

Was this helpful?

Now we get to the fun part - Training a custom model!

In this example we will be training a custom women's handbag model.

You can of course train any custom model you like after this workshop, but for this example we will take a look at retail and women's fashion!

Step 1:

To Train a custom model the first thing we need is some data.

Download the following files :

Finally the Test file to test our model

When you go off and create your own custom model you will need to find the data and create zip files to train.

Step 2 :

Go ahead and re name your classifier . I named mine Hand-Bag Classifier . Feel free to name your model anything you like depending on the type of classifier you are building . Also click browse to upload the chanel.zip , louis-viton.zip , and the dior.zip files into the project.

Step 3:

Once your files are uploaded you can click on the blue squares next to each zip file and click on Add to model

You should see something like this :

You might get a flag on the class name. Make sure to chage it so includes correct characters.

Doing this step ensures that you have three different classes of images in place.

Step 4 :

Click on Train in the upper right hand corner to train the model . Training can take 10 - 15 mins depending on model size.

Now all you have to do is wait for training to be done and then you can go ahead and test!

4MB
chanel.zip
archive
4MB
dior.zip
archive
dior.zip
3MB
louis-vitton.zip
archive
louis-vitton.zip
3MB
test.zip
archive
test.zip
You should see this message when you start training