Once you have framed your problem you can start to prepare the data. Our software is quite flexible already and takes several datatypes, such as images, PDF, text or sound. Depending on the data type, the input needs to be prepared in a slightly different way – here is how.
For all: More is better!
In order to train a machine, you need to show it some examples of what you want it to look at (data) and what it is supposed to predict (label). Our software allows you to upload both labeled and unlabeled data. However, prior to starting the training process, this data needs to be labeled. This can be done in different ways, e.g. Slack or Excel.