Please note that Neuton no longer supports Internet Explorer.
We recommend upgrading to the latest Microsoft Edge, Google Chrome, or Firefox.
Close
Neuton Quick Start Guide
LOG IN TO NEUTON
Go to https://neuton.ai in your web browser to access Neuton (you can skip the video by clicking «Try Neuton»). After page refresh click «Sign in with Google» to start using Neuton. You will be redirected to the «Sign in with Google» web page, after that please provide the email address and password used during Neuton registration.
SELECT DATA FOR TRAINING
Create your first solution using the «Add Solution» button.


Specify the «Solution Name» in the pop-up window.


Press the «Next» button to select the preloaded dataset or upload a new one.
On the next screen press the «Select Dataset» button. Navigate to the «Preloaded dataset» tab and choose Titanic_training.csv:

Training dataset should be a csv file with UTF-8 encoding. The first row in the training dataset must contain the column names and a comma must be used as a separator.

Press the «OK» button to confirm selected dataset.
Select Target Variable.

Press the «Next» button to go to next step.
TRAIN YOUR MODEL
On this step Neuton will automatically offer you Task type and target Metric but you can change it according to your preferences. To continue press «Start Training».


You can track the percentage of training completion on the progress bar at the top of the training tab.


As a result of training, the model is created.


Press the «OK» button to go to Prediction stage.
MAKE PREDICTIONS
When the training is completed click the «Enable» button to switch on prediction.


That additional costs may be charged for a new virtual machine for predictions.

When the Virtual machine is created, the «Start» button will be activated, press it to start Web Prediction and upload your test dataset (or new data) without target variable to make the prediction (this process is similar to dataset selection described for the training stage). You can download test dataset for preloaded dataset «Titanic_training.csv» from from the Kaggle Titanic ML competition.

The test dataset structure and variable types should be the same as in the training dataset, except that the target column will be excluded.

Wait for predicting completion. It can take a few minutes.


Congratulations! Your predictions are completed!


Download prediction results for further insights!
For more information about Data Science and improving prediction results on your dataset please refer to Neuton User Guide.
Contact Us
Get in touch to learn more