September 16, 2021

Models explainability, transformers, and automated data preparation

Model explainability

Try out the new model explanation dashboard. It provides the possibility to explain the outcome of the classification or regression model using the Shapley Additive Explanation approach.

Automatic Data Preparation

From now you can do the initial data preprocessing within a single dashboard without the necessity to configure the complex multi-brick pipeline. Try out the new Data Preparation brick that allows configuring the data cleansing and transformation scenario and immediately apply it to the dataset. The editing dashboard goes along with a detailed data summary for the input and transformed datasets.

Query preview in the data-sources

Now you can see the result of your SQL query


You can now reuse any data transformation rules learned and applied in the predictive pipelines. Try out the new transformers that currently have the models-like behavior.

Snowflake connector

Use the new Snowflake data connector to easily import your data into the platform.

Linked Models

No need to manually re-deploy the models anymore. Just link the training pipeline and use the latest version of your model.

Check out our previous updates

Datrics Updates: Better UX, New data uploader and Auto ML

Better UX, New data uploader and Auto ML

Added TimeSeries Forecasting brick which supports stratification and may be used for the time series forecasting without complex settings.
Datrics Updates: Better UX, New data uploader and Auto ML

Datrics $1M Seed announcement

We are happy to announce has raised a $1+ million seed round!
Datrics Updates: Governance, models reusability, outliers detection

Governance, models reusability, outliers detection

Better UI for models, now you can train multiple models inside for-loop and pipeline and assess their performance via quality metrics and diagrams.
Do you want to discover more about Datrics?