Deepomatic Platform

Training Models and Model Versions

Models in Machine Learning correspond to specific solutions, specific ways to address a problem. Within a project, a model then corresponds to a specific set of concepts. Adding or deleting a concept changes the model that you are building because it implies that you are modeling the problem differently, or creating a different solution to the same problem. However, adding new images to the project only adds data but does not alter the model.
Model Versions are instances of a solution to your problem. Within a project, a model version then corresponds to a specific trained neural network, with a specific architecture, and on a specific set of training images.

Models Library

When on a given project, to train model versions and to see training history, click on the Library tab in the Models section of the navigation bar. You will see a listing of all the models that have been created, the different versions that you have trained for each one of them, their status, a first indicator on their performances and a few useful links.

Train a model version

To train a new model version, click on Train a new model version, give a name to your model version, and decide on your parameters in the training options panel. By clicking on Create, you launch the training.
In the same way as the Add images, once you start a training, you will get a progress bar with the different steps required and a status on their advancement.
For more information on the advanced options panel, see our Guidebook on how to build your custom AI. For information on all the available models, click on the link below.
Once the training is launched, you can click on the model version to get more information on the training information and evaluate the performance of your model.

Training options

There are a few parameters you need to know about:
  • Architecture: The type of neural network that the model will use. You can select from a variety of state-of-the-art architectures whether you prefer to focus on accuracy or on computation speed.
  • Iterations: How long the training will last or more precisely, the number of passes (one pass corresponding to the forward of images into the neural network and the back propagation of the error in the neural layers) through the neural network. For each pass, the number of images that will be used is defined by the batch size. It can be found in the page Available architectures.
  • Learning rate: The strength of the learning, a higher rate means coarser learning while a lower rate indicates finer learning.
  • Optimizer: The learning strategy used for the neural network, works hand-in-hand with the learning rate.
  • Class balancing: Whether to use more often the concepts with fewer images during training. Used to correct the bias when a concept has very little images while another one has a lot.
The number of iterations set by default for some architecture may be oversized if you have few images in your dataset