Machine learning workflow — Knowledge check

Choose one answer per question. Submit your answers to see your score and an explanation for each answer.

Lesson 04 / From data to deployment0%

1. Watch the video2. Complete the test

Checking your lesson progress…

1 What is the first step in Neural Designer’s machine learning workflow? Train the neural network. Select the model type, such as regression, classification, or forecasting. Export the trained model.

2 What are the roles of the training, selection, and testing subsets? Training teaches the model, selection guides hyperparameter tuning, and testing evaluates final performance. Training evaluates final performance, while testing is used to adjust the model’s weights. All three subsets have the same purpose and can be used interchangeably.

3 What do you choose when configuring the training strategy? The report layout and the model’s export format. Only the number of rows and columns in the dataset. The loss function, optimization algorithm, and stopping criteria.

4 What is the goal of model selection? Choose a model using only its training error. Compare model configurations on the selection subset to reduce overfitting and choose a model that generalizes well. Replace the data preparation and training steps.

Next lesson →4 questions · Pass: 5.0/10

Return to the lesson

No-code AI Platform.

Find Us At: Calle del Adaja 10, 37185 Villamayor, Salamanca (Spain)

Company
Contact us
F.A.Q.
Legal notice
Cookie policy
(+34) 651 46 26 38
info@neuraldesigner.com
LinkedIn
X