Citrine v2.42.2 is released!

This release of Citrine Python includes only a minor bug fix to correct our deployment to PyPI.

Fix

• Fix to our automatic deployment to PyPI. #911

Coming Soon

• We are very excited to announce that Citrine python v3.0 is expected to release in early February 2024. Note this means that currently deprecated methods will result in errors after this move, so be sure to check your code for any deprecation warnings. Contact Citrine support for any assistance in migration.

Full Changelog: v2.42.1...v2.42.2

Citrine v2.42.1 is released!

In this version of Citrine python, we have some internal updates as well as improvements to our methods for accessing branches to improve consistency.

We are also pleased to announce that Citrine python v3.0 will be coming in early February 2024. So as you update to the latest version of Citrine python, be sure to check your code (especially automated scripts) for any deprecation warnings! Those deprecated methods will result in errors after moving to v3.0.

Improvements

• Improvements to internal test automation. #906

Deprecated

• Deprecate branch access by version ID. To remain consistent with our other methods, one instead gets a branch with the branch root ID and version number. #902

Coming Soon

• We are very excited to announce that Citrine python v3.0 is expected to release in early February 2024. Note this means that currently deprecated methods will result in errors after this move, so be sure to check your code for any deprecation warnings. Contact Citrine support for any assistance in migration.

Full Changelog: v2.41.0...v2.42.1

Citrine v2.41.0 is released!

Shipping fixes and internal improvements with this release of Citrine Python. Making sure you can run smoothly as we pave the way for the next big step in functionality.

Improvements

• Internal changes for feature development. #891, #892, #893, #894, #898

Fixes

• Add admin checks for users and teams. #901

Full Changelog: v2.37.1...v2.41.0

Citrine v2.37.1 is released!

Some minor updates to Citrine Python in this release to help catch ingestion errors and keep up with changing dependencies.

Improvements

• Add timeout setting to files.ingest ingestion endpoint to catch errors. #889
• Maintenance to keep us secure and up to date with dependencies. #890

Full Changelog: v2.36.0...v2.37.1

Citrine v2.36.0 is released!

This release of Citrine python is a bit sleepy but powerful nonetheless. We've continued to improve our error communication in the data ingestion process. We've also made internal improvements for our developers and to pave the way for future functionality.

Improvements

• New error surfaced when ids collide during ingestion. #884
• Minor internal improvements. #885, #886, #887

Full Changelog: v2.34.0...v2.36.0

Citrine v2.34.0 is released!

In this release of Citrine Python, we're introducing a host of improvements to address the feedback of our users and prepare for new functionality. As always, we hope to keep you running smoothly.

Improvements

• Fields are now optional in the predictor.rename method. #878
• We've broadened the acceptable arguments of ingest file names to accept path-like arguments. #880
• Improvements to consistency in error handling during ingestion. #882
• Minor internal improvements. #879, #876

Deprecated

• Deprecation warning added for descriptors.descriptors_from_data_source in favor of descriptors.from_data_source. #881

Full Changelog: v2.28.2...v2.34.0

Citrine v2.28.2 is released!

In the latest release of Citrine python, we have some new alpha functionality along with improvements and a fix. Our backend now supports integers as descriptor types in the AI Engine (they have long been supported in our data model), though that is considered alpha until our web application UI fully supports it. We've also improved our responses returned from ingestion tasks for easier de-bugging, and improved our documentation. As always, we are continually fixing bugs that arise and improving our backend to keep you running smoothly.

What's New

• [ALPHA] We now support integers in our AI Engine! Citrine python users can leverage the integer descriptor and integer constraints for modeling and design. However, this will cause instabilities in the UI of the Citrine Platform web application in the near term. Expect UI errors until cleared from ALPHA. #873, #877

Improvements

• Improvements to how our ingestion workflow returns statuses and errors to our users. #869
• Improve documentation on file upload. #875

Fixes

• Fix error default design space creation. #874

Full Changelog: v2.26.1...v2.28.2

Citrine v2.26.1 is released!

In this release of Citrine Python, we've added a few small but key additions. You can now edit a Predictor without having to wait for a retrain. Also, we've added documentation on how to use the sample_design_space feature to return a list of starting candidates for your SL project, even if you don't have any data. And of course, we have some minor changes to keep up with our evolving back-end.

What's New

• You can change a Predictor configuration without triggering a retrain by passing train=False when updating your Predictor. #867

Improvements

• Add documentation on sampling a Design Space. #870
• Updated design spaces to use new endpoints. #868, #871

Full Changelog: v2.24.1...v2.26.1

Citrine v2.24.1 is released!

With this release we are excited to provide two new features for our users. You can now quickly change the name of a predictor without re-training and you can get a list of experiments even if you haven't yet built a model! As always, we've kept up our maintenance to keep you running smoothly.

What's New

• You can now change the name and/or description of a predictor without triggering a full validation and retraining process. Simply call project.predictors.rename(uid, version, name, descriptor). #865
• Want to kick start your experiments but don't have enough data for a model? Now all you need to is a Design Space! Triggering a Sample Design Space Execution will give you a diverse list of candidates to take to the lab - no model needed. #860

Improvements

• Minor improvements for our developers. #861, #862, #864, #866

Full Changelog: v2.22.0...v2.24.1

Citrine v2.22.0 is released!

We are excited to announce some highly impactful new features with this release of Citrine Python. You can now get the fully featurized training data associated with any valid predictor. We've also included more control over the generative molecular design process, as well as improved the data ingestion process.

What's New

• Finally, you can quickly access the fully featurized training data from a given predictor with the get_featurized_training_data method. Simply call project.get_featurized_training_data with the ID and version of your predictor and you will get a list of design materials representing your training data with features present. #858
• You can now designate the minimum number of substructures to construct when running generative molecular design by passing a dictionary to the new min_substructure_counts argument. #856
• Thanks to #859, a slew of new arguments to the ingest method make getting structured data on platform substantially easier:
  • upload will copy files onto platform if they're not there already
  • delete_dataset_contents and delete_templates will clear out old records prior to creating new ones
  • build_table will follow object generation with creating a table for that data, ready for modeling

Improvements

• FileLink objects and the associated FileCollection methods (read, write, etc.) now work smoothly with local files just like remote ones: just use the File URI scheme or convenient from_path method. #859

Full Changelog: v2.19.2...v2.22.0