-
Unique key detector tool
What's your use case? More than often, you have to deal with a dataset without knowing what's make a row unique. This can lead to misinterpret the data, cartesian product at join and other funny stuff. What's your proposed solution? This is a feature I haven't seen in any data prepation/etl. The core feature is to detect…
-
ADBC connectivity : faster columnar storage query
Hello all, ADBC is a database connection standard (like ODBC or JDBC) but specifically designed for columnar storage (so database like DuckDB, Clickhouse, MonetDB, Vertica...). This is typically the kind of stuff that can make Dataiku way faster. more info in https://arrow.apache.org/blog/2023/01/05/introducing-arrow-adbc/…
-
Convert Visual and Structured Visual Agent to Agent Code
Hello, Visual and Structured Visual Agents are easy to design, a lot easier than an agent code. However, it creates a lot of dependancy to Dataiku, meaning some restrictions to use them. On the other side, a pure code agent is pretty hard to desig but it brings some freedom. I tried to convert a SV agent to code agent,…
-
Dataiku Headless: Add Support for SSO Authentication
As industry moves towards passwordless, this is a request to provide support for SSO-based authentication for Dataiku Headless (OAuth Authorization Code flow with PKCE), as an alternative to requiring users to manually create, configure and manage a Dataiku API keys. Users should be able to authenticate interactively…
-
Improve integration of custom and plugin models in VisualML
The functionality to continuously expand VisualML with plugin models, either from Dataiku or by coding our own plugins is very valuable to us. For example, we use the General Linear Models plugin a lot to support our actuarial team. There are some areas where the integration of such plugins is not yet ideal in VisualML…
-
Labeling > Support providing Annotations as optional Input
Hi, I am using the new Labeling recipe as the name suggest to add labels to unlabelled images. In my use case, the images are quite big in terms of dimensions and each images requires around 10-20 annotations. This means, the manual task of labeling a decent amount of images is quite high. So far I have labelled around 70…
-
Visual ML + Deployed Model Diagnostic Outputs
It would be great to be able to export data related to failed or ignored rows within both the visual ml and deployed model diagnostics with a label denoting the reason for not training. This would keep the user from using other methods to a. locate which rows did not get trained and b. the reason for them not getting…
-
Retraining Partitions for Partitioned Time Series Modeling
Currently when aborting a partitioned time series model you are able to "rerun" one model on the aborted partitions however seem to be unable to rerun all aborted partitions within a session. For example: Training a partitioned model using seasonal naive and trivial identity in a single session and abort the session. I can…
-
Default settings for importing datasets
So, I literally never want my integers coerced to doubles, and it would be great to have the option to have the "Normalize floats & doubles" box default to unchecked when I'm creating a new dataset. Maybe on a per-project basis? Maybe a Datasets section in the Project Settings? Similarly, I basically always want my dates…
-
Multi-Task Feature
Hi/Hello from Matraman Indonesia. I’m trying to understand whether Dataiku Visual ML currently supports a true multi-target prediction task beyond multi-target regression. From the documentation, I can see that Multi-target Regression allows multiple numerical targets. However, I’m wondering about a case where I have…