v0.0.16

What's Changed

• Fix example selector with non string inputs by @rajasbansal in #558
• Change label to null label on output not followed by @rajasbansal in #559
• Increase pool size and record prompts after labeling by @rajasbansal in #549
• Support disable quoting while reading .csv files by @mkchaitanya03 in #563
• Support full named-entity level metrics by @mkchaitanya03 in #562
• Add an attribute extraction task by @rajasbansal in #566
• Add support for Redis Cache by @yadavsahil197 in #570
• Better support for classification tasks with large number of label classes by @iomap in #561
• Allow confidence computation per attribute and setting label columns by @rajasbansal in #572
• Change embedding model attribute's name in vector store wrapper by @mkchaitanya03 in #580
• Expanding benchmark by @DhruvaBansal00 in #584
• Allow additional properties by @Abhinav-Naikawadi in #585
• add documentation for using label_selection on large classification tasks by @iomap in #586
• Anthropic bugfix - upgrading sdk, tokenizer usage by @DhruvaBansal00 in #597
• Bug fixes and add documentation for attribute extraction by @rajasbansal in #596
• Refuel LLM integration with autolabel by @rajasbansal in #595
• add initialize to cache by @rajasbansal in #598
• bump version for release by @rajasbansal in #601

New Contributors

• @mkchaitanya03 made their first contribution in #563

Full Changelog: v0.0.15...v0.0.16

v0.0.15

What's Changed

• Fix AUROC Bug and allow custom Example Selectors by @rajasbansal in #531
• Setting skip keys as true for hashing non basic types by @rajasbansal in #532
• Make task management optional by @rajasbansal in #535
• Update example notebooks with the new agent changes by @nihit in #536
• optionalize evaluation and console progress printing by @nihit in #537
• typo in getting started docs by @guptav96 in #538
• Transforms: Send only specified output columns in transformed row by @yadavsahil197 in #539
• Synthetic Data Generation With LLMs by @Tyrest in #500
• Add annotations column in output dataframe by @Abhinav-Naikawadi in #541
• Transforms: Add error handling by @yadavsahil197 in #540
• Fix error message on openai key not set by @rajasbansal in #543
• added ocr image transform by @Tyrest in #515
• fix bugs for kill signal and indexing by @Tyrest in #545
• added docs for synthetic datasets and progress by @Tyrest in #544
• fix some outdated examples in the documentation by @nihit in #548
• Add support for Serp Api Transform by @yadavsahil197 in #547
• Update name of db with the new release by @rajasbansal in #551
• bump version for release by @nihit in #557

New Contributors

• @guptav96 made their first contribution in #538

Full Changelog: v0.0.14...v0.0.15

v0.0.14

What's Changed

• Allow label column to be null for running autolabel by @rajasbansal in #485
• Bump version to 0.0.10 by @rajasbansal in #488
• Autolabel config cli by @Tyrest in #463
• cli file name bugfix by @Tyrest in #493
• added logit bias for huggingface by @Tyrest in #495
• Added configurable log level to the CLI by @Tyrest in #497
• CLI Logit Bias and Model Params Fix by @Tyrest in #498
• Output object instead of dataframe to handle sorting and filtering by @rajasbansal in #482
• Add transform base class with a very simple agent.transform by @rajasbansal in #496
• added CLI documentation by @Tyrest in #494
• added links to the CLI documentation by @Tyrest in #505
• updated cli to reflect the transform change by @Tyrest in #504
• Accept config in LabelingAgent by @rajasbansal in #501
• Add support for Webpage Transform by @yadavsahil197 in #507
• added pdf transform and tests by @Tyrest in #502
• Support custom llm by @Sardhendu in #503
• Add cache for transforms and unpin sqlite by @rajasbansal in #508
• fixes to cli documentation like lists and others by @Tyrest in #514
• remove data loader directory (unused) by @rajasbansal in #512
• Update docs and add an example webpage transform notebook by @rajasbansal in #517
• added pdf transform documentation by @Tyrest in #519
• Allow running transforms in Jupyter Notebook using nest_asyncio by @rajasbansal in #520
• Add docs for webpage transform by @rajasbansal in #521
• version bump by @nihit in #522
• Bump version and fix dependency issue by @rajasbansal in #524
• edited pdf transform documentation for consistency by @Tyrest in #523
• upgrade bs4 version by @nihit in #525
• bump version by @nihit in #526
• Update docs for Autolabel Dataset by @rajasbansal in #527
• Add optional vector store params to autolabel config by @Abhinav-Naikawadi in #528
• fix config type error in AutolabelDataset initialization by @nihit in #530

Full Changelog: v0.0.9...v0.0.14

v0.0.13

What's Changed

• Allow label column to be null for running autolabel by @rajasbansal in #485
• Bump version to 0.0.10 by @rajasbansal in #488
• Autolabel config cli by @Tyrest in #463
• cli file name bugfix by @Tyrest in #493
• added logit bias for huggingface by @Tyrest in #495
• Added configurable log level to the CLI by @Tyrest in #497
• CLI Logit Bias and Model Params Fix by @Tyrest in #498
• Output object instead of dataframe to handle sorting and filtering by @rajasbansal in #482
• Add transform base class with a very simple agent.transform by @rajasbansal in #496
• added CLI documentation by @Tyrest in #494
• added links to the CLI documentation by @Tyrest in #505
• updated cli to reflect the transform change by @Tyrest in #504
• Accept config in LabelingAgent by @rajasbansal in #501
• Add support for Webpage Transform by @yadavsahil197 in #507
• added pdf transform and tests by @Tyrest in #502
• Support custom llm by @Sardhendu in #503
• Add cache for transforms and unpin sqlite by @rajasbansal in #508
• fixes to cli documentation like lists and others by @Tyrest in #514
• remove data loader directory (unused) by @rajasbansal in #512
• Update docs and add an example webpage transform notebook by @rajasbansal in #517
• added pdf transform documentation by @Tyrest in #519
• Allow running transforms in Jupyter Notebook using nest_asyncio by @rajasbansal in #520
• Add docs for webpage transform by @rajasbansal in #521
• version bump by @nihit in #522
• Bump version and fix dependency issue by @rajasbansal in #524
• edited pdf transform documentation for consistency by @Tyrest in #523
• upgrade bs4 version by @nihit in #525
• bump version by @nihit in #526

Full Changelog: v0.0.10...v0.0.13

v0.0.9

What's Changed

• add multilabel classification to validation by @Tyrest in #445
• consolidate compute f1 to single function in utils by @Tyrest in #453
• cache embeddings in vector store by @Tyrest in #457
• Add llm explanation to output dataframe by @rajasbansal in #462
• Rmv chunk size by @Sardhendu in #459
• added output cost calculation to PaLM get_cost by @Tyrest in #470
• Fix metric calculation for unlabeled datapoints by @rajasbansal in #468
• fixed adding cache argument to certain algorithms by @Tyrest in #471
• Added causal LM support by @Tyrest in #472
• added causal lm documentation by @Tyrest in #477
• Bubble up errors and clean up error creation by @rajasbansal in #475
• Metric calculation as a separate class by @rajasbansal in #480
• version bump + clean build artifacts by @nihit in #483

Full Changelog: v0.0.8...v0.0.9

v0.0.8

What's Changed

• Fix 178: Add data type checking and validation while performing agent plan action by @Sardhendu in #428
• Support example selection using label diversity by random and by semantic similarity by @Abhinav-Naikawadi in #436
• rmv duplicate and add adjustments to ledger config by @Sardhendu in #437
• Correcting name of 16k context model for OpenAI LLM docs by @rishabh-bhargava in #439
• label diversity example selection docs by @Abhinav-Naikawadi in #438
• Format seed file path correctly by @Abhinav-Naikawadi in #442
• add support for older pinned gpt-3.5-turbo-0301 by @nihit in #444
• Dataset validation fix for explanation generation by @Abhinav-Naikawadi in #441
• Documentation updates for few-shot and LLMs by @nihit in #446
• added individual retrying for failed batches by @Tyrest in #443
• add chain_of_thought flag to the config schema by @nihit in #447
• Revise guidelines and examples for chain of thought by @nihit in #448
• tag sections in the civil comments notebooks by @nihit in #450

Full Changelog: v0.0.7...v0.0.8

v0.0.7

What's Changed

• added cohere support by @Tyrest in #419
• Autolabel embedding config docs by @Abhinav-Naikawadi in #416
• Add docs for cohere in llms.md by @Tyrest in #420
• Bump version: 0.0.6 -> 0.0.7 by @rishabh-bhargava in #426
• Multi label docs visibility on docs.refuel.ai by @rishabh-bhargava in #427
• Fix open api key error handling by @chirag-manwani in #423
• Cohere embeddings support by @Abhinav-Naikawadi in #432

New Contributors

• @chirag-manwani made their first contribution in #423

Full Changelog: v0.0.6...v0.0.7

v0.0.6

What's Changed

• Configs docs by @Abhinav-Naikawadi in #320
• change readme links for readability by @nihit in #322
• revamp tutorial in the docs website by @nihit in #326
• Bump version by @rishabh-bhargava in #327
• typeform link in readme by @nihit in #329
• update the image format in README by @nihit in #331
• merge assets and test/assets directories by @Abhinav-Naikawadi in #330
• Bump version: 0.0.2 -> 0.0.3 by @rishabh-bhargava in #332
• add link to form for confidence api key by @Abhinav-Naikawadi in #333
• Add remaining colab integrations, fix docs deprecated configs, and some example notebook cleanups by @Abhinav-Naikawadi in #335
• Chain of thought json parsing by @rajasbansal in #336
• Fix the labeled stats shown for labeled examples by @rajasbansal in #334
• put links next to section headers in docs site by @iomap in #337
• add support for new OpenAI models, update pricing by @nihit in #339
• Adding single quotes around pip install in CONTRIBUTING.md by @rishabh-bhargava in #341
• Bump version for chain of thought fixes by @Abhinav-Naikawadi in #338
• remove the -inf row with compute confidence by @Tyrest in #343
• add support for all gpt-3.5 and gpt-4 model versions by @nihit in #346
• added retry with tenacity for palm by @Tyrest in #347
• Adding a helpful video during set up for contributing by @rishabh-bhargava in #348
• Check if the LLM label is in the labels list by @Tyrest in #345
• update Makefile help output by @nihit in #353
• Moving returned items to dataset loader fields by @rajasbansal in #317
• Update get_data and add test case by @Sardhendu in #351
• Fixing a small typo in the confidence section of docs by @rishabh-bhargava in #358
• Update llms.md by @gruentee in #359
• Update llms.md by @eltociear in #360
• added custom progress bar for get data by @chiranthans23 in #361
• fix test failures by @nihit in #364
• Adding single quotes around pip install extras by @rishabh-bhargava in #363
• add pull request trigger to actions by @nihit in #365
• Adding default cpu support. Removing max tokens for huggingface warnings by @DhruvaBansal00 in #366
• Revamp civil comments example notebook and add Colab links by @nihit in #367
• added links for convenience to llms.md by @JTarakRam in #369
• Fix JSON config for starter task by @yadavsahil197 in #372
• Add mkdocs build dependencies to dev setup by @yadavsahil197 in #373
• Checks should run on PR commits and after push to main branch by @yadavsahil197 in #375
• Pin Langchain to a working version by @yadavsahil197 in #374
• added support for hugging face datasets by @Tyrest in #354
• version bump by @nihit in #376
• added small hugging face dataset section to docs by @Tyrest in #377
• added comma's to fix syntax errors in llms.md by @JTarakRam in #381
• Fix 196 by @Sardhendu in #384
• small fixes to config validation by @nihit in #386
• revamp roadmap section by @nihit in #402
• pin anthropic version to avoid breaking changes by @Tyrest in #407
• Support for multiple embedding providers (Huggingface, etc.) by @Abhinav-Naikawadi in #404
• Add multi label classification task by @Tyrest in #405
• Fix engine path by @turian in #414
• Add docs for multilabel classification by @Tyrest in #411
• added check for refuel api key in plan by @Tyrest in #344
• Black formatting checks to run on PR branch by @yadavsahil197 in #417
• Implemented logit bias by @Tyrest in #385
• Bump version: 0.0.5 -> 0.0.6 by @rishabh-bhargava in #418

New Contributors

• @gruentee made their first contribution in #359
• @eltociear made their first contribution in #360
• @chiranthans23 made their first contribution in #361
• @JTarakRam made their first contribution in #369
• @turian made their first contribution in #414

Full Changelog: https://github.com/refuel-ai/autolabel/commits/v0.0.6