Getting started with No-code ML

The No-code ML module of TOKII is a machine learning module that allows users to create, train, and deploy predictive models without the need for programming knowledge. Any user can experiment with these approaches in a guided, visual, and intuitive way. This democratizes the use of ML and allows professionals from different areas to incorporate predictive capabilities into their processes easily and efficiently.

Based on the datasets connected or uploaded in TOKII (link), the No-code ML module allows for a complete training flow in just a few steps:

🔧 Create a training

Start a new training session from the AI module by assigning a name and description. Select the previously uploaded dataset in TOKII and define the type of learning: supervised (classification or regression) or unsupervised (clustering). Then choose the corresponding algorithm, such as decision tree, neural network, or K-means, as appropriate for the case.

⚙️ Configure the training

Define the input variables, the objective (if applicable), the test percentage, and adjust the algorithm parameters. This configuration determines how the model will behave when trained.

🔁 Test and repeat: create and run new sessions

Run the training, analyze the results, consult metrics and graphs generated automatically. You can create multiple sessions with different configurations to compare and improve the performance of your models.

🚀 Deploy the best model

Use the most suitable model to generate predictions on new data. These predictions can be used directly or displayed in real-time in TOKII's digital twin KPIs to support operational or strategic decisions.