v0.6.4

New Features

• feat: support custom and integer frequencies @jmoralez (#532)
• feat: usage method @jmoralez (#548)
• feat: add hist_exog_list argument to cross_validation @jmoralez (#534)

Documentation

• Add Why TimeGPT notebook @Yibei990826 (#537)
• [FIX] Pricing links incorrect @elephaint (#544)
• [FIX] Update contributing guide docs @elephaint (#539)
• fix: cleaned the FAQ related to API validation @ngupta23 (#538)
• docs: instructions on how to contribute @ngupta23 (#526)
• docs: broken links in intro notebook @ngupta23 (#527)
• docs: updated dashboard image to latest dashboard @ngupta23 (#528)

v0.6.3

Bug Fixes

• fix(deps): bump utils @jmoralez (#524)

v0.6.2

New Features

• FEAT: Add finetune_depth parameter @marcopeix (#471)

Breaking Change

• breaking: raise error for gaps in series @jmoralez (#504)
• breaking: add hist_exog_list argument to forecast @jmoralez (#505)

Bug Fixes

• fix: convert level to quantiles in historic forecast @jmoralez (#510)
• fix: reduce test time @elephaint (#487)
• fix(ci): increase tolerance in tests @jmoralez (#503)
• fix: update docs to add hist exog list instead of previous behaviour @AzulGarza (#520)

Documentation

• Updating Azure getting started guide. @tracykteal (#508)
• updating Azure and trial information @tracykteal (#500)
• Adding a pricing page to our docs @tracykteal (#502)
• docs: add finetune_depth to tutorial on improving accuracy @marcopeix (#497)
• docs: add improve accuracy content to special topics page @Yibei990826 (#496)
• docs: add notebook tutorial for improve forecast accuracy @Yibei990826 (#495)
• docs: update energy demand forecast tutorial @Yibei990826 (#499)

Enhancement

• enh: warn when overriding model in Azure endpoint @jmoralez (#511)
• enh: improve short series error message @jmoralez (#494)

v0.6.1

New Features

• feat: support newer azure deployments by @jmoralez in #478

Documentation

• [DOCS] Add API call count FAQ by @marcopeix in #449
• Adding Azure getting started doc by @tracykteal in #474

Other

• chore: remove autogenerated code by @jmoralez in #468

Full Changelog: v0.6.0...v0.6.1

v0.6.0

Breaking changes

• Removed TimeGPT class in favor of NixtlaClient.
• Removed NixtlaClient.validate_token method in favor of NixtlaClient.validate_api_key.
• Removed short-horizon and long-horizon models in favor of timegpt-1 and timegpt-1-long-horizon respectively.
• Removed fewshot_steps and fewshot_loss in favor of finetune_steps and finetune_loss respectively.
• Removed TIMEGPT_TOKEN environment variable in favor of NIXTLA_API_KEY.
• Timestamps in the time_col preserve their type in the outputs (timestamp), previously they were cast to string.
• Gaps in series are not filled anymore, in line with our documentation. Please use the fill_gaps function if you require to do so.
• NixtlaClient.weights_x is now a list of lists if num_partitions != None, where each element corresponds to the weights for a specific partition.

Features

• feat: call v2 endpoints by @jmoralez in #429
• feat: orjson serialization by @jmoralez in #452
• [FEAT] Add historical exogenous by @elephaint in #453
• FEAT: Add feature contributions as attribute and tutorial on how to use by @marcopeix in #448
• [FEAT] - Raise error when payload is too large and suggest number of partitions by @marcopeix in #456

Bug fixes

• fix: removed data input restriction during cross validation finetune by @Yibei990826 in #426

Enhancements

• feat: use TypeVar for dataframes by @jmoralez in #443
• feat: set supported_models using base_url by @jmoralez in #450
• feat: suggest earlier version for azure endpoints by @jmoralez in #457

Documentation

• updating excel docs with some more information by @tracykteal in #414
• updating FAQ for fee information and Azure being available by @tracykteal in #412
• Update 03_excel_addin.ipynb by @tracykteal in #416
• [DOC] - Rephrase the sentence for specific loss function by @marcopeix in #436
• docs(readme): getting started by @mergenthaler in #438
• Adding activate trial link in intro by @tracykteal in #417
• [DOC] - Add instructions on saving figures when not working in notebooks by @marcopeix in #454
• [BUG]: close img tag in introduction notebook by @marcopeix in #441
• docs: add polars quickstart by @jmoralez in #447

Full Changelog: v0.5.2...v0.6.0

v0.5.2

What's Changed

Features

• Use case on missing values by @MMenchero in #326
• Electricity demand use case by @marcopeix in #321
• [FEAT] What if - pricing in retail scenario by @elephaint in #340
• Update the middleware endpoint to use the new one by @loama in #407
• Update excel add in docs to reflect changes in the plugin by @loama in #406
• feat: improve readability by @mergenthaler in #367
• [DOC] Add azure callouts to use cases by @elephaint in #357
• [DOC] - Add links and callouts by @marcopeix in #355
• [DOC] Model callouts and descriptions by @MMenchero in #359
• feat: add foundation ts arena by @AzulGarza in #378
• feat: Added nixtlar to TimeGPT docs by @MMenchero in #393

Fixes

• [FIX] SDK Reference by @elephaint in #342
• [FIX] Raise warning on missing X_df when df has exogenous by @elephaint in #295
• fix: add jupyter lab to dev deps by @mergenthaler in #350
• [FIX] Hotfix for old SDK link removal by @elephaint in #352
• [Hotfix] Minimum data requirements by @MMenchero in #364
• fix: note on morai improvements by @mergenthaler in #395
• fix: Clarify how tokens are counted for AzureAI by @MMenchero in #401
• Hotfix for links in documentation by @MMenchero in #341
• [FIX] Colab-flag by @elephaint in #349
• Minor correction to tutorial by @MMenchero in #394

Chore

• Readme update by @mergenthaler in #346
• Add Fern badge to README.md by @dannysheridan in #185
• [DOC] - Reduce code blocks in capabilties by @marcopeix in #339
• Better intro/welcome page for doc by @marcopeix in #343
• a few updates to getting started by @tracykteal in #344
• fix: note on chronos pr by @mergenthaler in #390
• fix: Change how data is downloaded for Bitcoin tutorial by @MMenchero in #391
• Removed a wrong sentence in README.md by @B-Deforce in #372

New Contributors

• @dannysheridan made their first contribution in #185
• @B-Deforce made their first contribution in #372
• @loama made their first contribution in #407

Full Changelog: v0.5.1...v0.5.2

v0.5.1

What's Changed

• [DOCS] - Add a welcome/introduction page to the documentation by @marcopeix in #319
• [FIX] Doc fixes by @elephaint in #324
• Add capabilities notebooks by @marcopeix in #312
• [DOCS] Fix structure by @elephaint in #329
• [FIX] Mintlify prefixes, readme file dirs by @elephaint in #333
• [FIX] Doc fixes by @elephaint in #334
• fix: add pyreadr as dev dep by @AzulGarza in #336
• fix: update distributed to computing at scale by @AzulGarza in #337
• fix: add model to cross validation path call by @AzulGarza in #335
• v0.5.1 by @AzulGarza in #338

Full Changelog: v0.5.0...v0.5.1

v0.5.0

What's Changed

New features

• regenerate fern code to support pydantic v1 and v2 by @jmoralez in #308
• convert ids to categorical by @jmoralez in #313
• feat: restrict input forecast by @AzulGarza in #314
• call corresponding endpoint in cross_validation by @jmoralez in #300
• store x_cols as a variable instead of attribute by @jmoralez in #318

Docs

• fix: update min version and nixtla_client by @AzulGarza in #317
• [FEAT] - Tutorials update (Marco) by @marcopeix in #311
• [DOCS] Add tutorials by @elephaint in #306
• New versions of how-to-guides for computing at scale by @MMenchero in #302
• Updated tutorials by @MMenchero in #316
• [DOCS] New structure by @elephaint in #323

Full Changelog: v0.4.0...v0.5.0

v0.4.0

What's Changed

Python package name change

We're deprecating the nixtlats package in favor of nixtla. Please ensure you make the following changes:

1. Install the nixtla package: pip install nixtla
2. Update your imports accordingly, e.g. from nixtla import NixtlaClient

New features

• [FEAT] Process multiple partitions in parallel by @elephaint in #296

Enhancements

• Create CODE_OF_CONDUCT.md by @tracykteal in #292
• use fill_gaps in resample_dataframe by @jmoralez in #267

New Contributors

• @tracykteal made their first contribution in #292
• @elephaint made their first contribution in #296

Full Changelog: v0.3.0...v0.4.0

v0.3.0

What's Changed

Changelog

0.3.0

🔄 Changes & Deprecations

• Deprecation of TimeGPT Class:
  In an effort to streamline our API and align with industry best practices, we're deprecating the TimeGPT class in favor of the new NixtlaClient class. This change is designed to provide a more intuitive and powerful interface for interacting with our services.

  Before:

  from nixtlats import TimeGPT
  
  # Initialize the TimeGPT model
  timegpt = TimeGPT()

  After:

  from nixtlats import NixtlaClient
  
  # Initialize the NixtlaClient
  nixtla = NixtlaClient()

• Renaming of Configuration Parameters:
  To enhance clarity and consistency with other SDKs, we've renamed the token parameter to api_key and environment to base_url.

  Before:

  timegpt = TimeGPT(token='YOUR_TOKEN', environment='YOUR_ENVIRONMENT_URL')

  After:

  nixtla = NixtlaClient(api_key='YOUR_API_KEY', base_url='YOUR_BASE_URL')

• Introduction of NixtlaClient.validate_api_key:
  Replacing the previous NixtlaClient.validate_token method, this update aligns with the new authentication parameter naming and offers a straightforward way to validate API keys.

  Before:

  timegpt.validate_token()

  After:

  nixtla.validate_api_key()

• Environment Variable Changes:
  In line with the renaming of parameters, we've updated the environment variables to set up the API key and base URL. The TIMEGPT_TOKEN is now replaced with NIXTLA_API_KEY, and we've introduced NIXTLA_BASE_URL for custom API URLs.

• Backward Compatibility & Future Warnings:
  These changes are designed to be backward compatible. However, users can expect to see future warnings when utilizing deprecated features, such as the TimeGPT class.

0.2.0 (Previously Released)

🔄 Changes & Deprecations

• Renaming of Fine-Tuning Parameters:
  The finetune_steps and finetune_loss parameters were renamed to fewshot_steps and fewshot_loss. Additionally, the model parameter values changed from short-horizon and long-horizon to timegpt-1 and timegpt-1-long-horizon, with an emphasis on preserving backward compatibility. In version 0.3.0, these changes are deprecated in favor of reverting to the original parameter names and values, ensuring a seamless transition for existing users.

Features

• feat: update docs in #268
• feat: add moirai benchmark in #271
• feat: replace token by api_key and environment by base_url in #275
• feat: replace TimeGPT class by NixtlaClient class in #276

Fixes

• fix: recover finetune and model changes in #274
• fix nixtla_client page render in #279

Full Changelog: v0.2.0...v0.3.0