Machine Learning (ML) on supervised learning (SL)

Machine Learning is using supervised learning to train a model on t6 IoT platform. t6 is using Tensorflow to fully handle ML

Tagged on #Machine-Learning, #Algo, #Supervised, #MadeWithTFJS, #TensorFlow, #WebML, #Classification, #Labelling,

ML Model definition

The first step in the supervised learning feature on the t6 IoT platform is model definition. This step involves specifying the architecture, parameters, and settings of the machine learning model that will be used for analysis. By providing a customizable interface through APIs, developers can easily define and configure their models to meet their unique requirements.

Technical Model Api doc

Training dataset

Once the model is defined, the next step is dataset training. The t6 IoT platform supports training models using labeled datasets, allowing users to teach the model to recognize patterns and make accurate predictions. Developers can capture real-time data from connected Objects to create comprehensive training sets. With this feature, users can train their models to accurately classify data, make predictions, and identify anomalies within their IoT data streams.

t6 provide robust support for both continuous and categorical features.

Technical Training Api doc

Predicting

The final step in the supervised learning feature on the t6 IoT platform is prediction. Once the model is trained, it can be used to infer predictions on new and unseen data points. The t6 IoT platform provides a seamless interface for users to feed new data to the trained model and obtain real-time predictions. These predictions can be used to make informed decisions, automate processes, or trigger appropriate actions based on the analyzed data.