From b28a4e97b937801fe65130df102cb628f534bb3c Mon Sep 17 00:00:00 2001 From: Ethan Harris Date: Tue, 15 Feb 2022 18:26:38 +0000 Subject: [PATCH] Bump version to 0.7.0 (#1174) --- CHANGELOG.md | 39 +-------------------------------------- flash/__about__.py | 14 +++++--------- 2 files changed, 6 insertions(+), 47 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 3f1c269dd2..65a1e6c3b5 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,94 +4,57 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/). -## [Unreleased] - YYYY-DD-MM +## [0.7.0] - 2022-15-02 ### Added - Added support for multi-label, space delimited, targets ([#1076](https://github.com/PyTorchLightning/lightning-flash/pull/1076)) - - Added support for tabular classification / regression backbones from PyTorch Tabular ([#1098](https://github.com/PyTorchLightning/lightning-flash/pull/1098)) - - Added Flash zero support for tabular regression ([#1098](https://github.com/PyTorchLightning/lightning-flash/pull/1098)) - - Added support for COCO annotations with non-default keypoint labels to `KeypointDetectionData.from_coco` ([#1102](https://github.com/PyTorchLightning/lightning-flash/pull/1102)) - - Added support for `from_csv` and `from_data_frame` to `VideoClassificationData` ([#1117](https://github.com/PyTorchLightning/lightning-flash/pull/1117)) - - Added support for `SemanticSegmentationData.from_folders` where mask files have different extensions to the image files ([#1130](https://github.com/PyTorchLightning/lightning-flash/pull/1130)) - - Added `FlashRegistry` of Available Heads for `flash.image.ImageClassifier` ([#1152](https://github.com/PyTorchLightning/lightning-flash/pull/1152)) - - Added support for `ObjectDetectionData.from_files` ([#1154](https://github.com/PyTorchLightning/lightning-flash/pull/1154)) - - Added support for passing the `Output` object (or a string e.g. `"labels"`) to the `flash.Trainer.predict` method ([#1157](https://github.com/PyTorchLightning/lightning-flash/pull/1157)) - - Added support for passing the `TargetFormatter` object to `from_*` methods for classification to override target handling ([#1171](https://github.com/PyTorchLightning/lightning-flash/pull/1171)) ### Changed - Changed `Wav2Vec2Processor` to `AutoProcessor` and seperate it from backbone [optional] ([#1075](https://github.com/PyTorchLightning/lightning-flash/pull/1075)) - - Renamed `ClassificationInput` to `ClassificationInputMixin` ([#1116](https://github.com/PyTorchLightning/lightning-flash/pull/1116)) - - Changed the default `learning_rate` for all tasks to be `None`, corresponding to the default for your chosen optimizer ([#1172](https://github.com/PyTorchLightning/lightning-flash/pull/1172)) -### Deprecated - ### Fixed - Fixed a bug when not explicitly passing `embedding_sizes` to the `TabularClassifier` and `TabularRegressor` tasks ([#1067](https://github.com/PyTorchLightning/lightning-flash/pull/1067)) - - Fixed a bug where under some circumstances transforms would not get called ([#1072](https://github.com/PyTorchLightning/lightning-flash/pull/1072)) - - Fixed a bug where prediction would sometimes give the wrong number of outputs ([#1077](https://github.com/PyTorchLightning/lightning-flash/pull/1077)) - - Fixed a bug where passing the `val_split` to the `DataModule` would not have the desired effect ([#1079](https://github.com/PyTorchLightning/lightning-flash/pull/1079)) - - Fixed a bug where passing `predict_data_frame` to `ImageClassificationData.from_data_frame` raised an error ([#1088](https://github.com/PyTorchLightning/lightning-flash/pull/1088)) - - Fixed a bug where segmentation files / masks were loaded with an inconsistent ordering ([#1094](https://github.com/PyTorchLightning/lightning-flash/pull/1094)) - - Fixed a bug with `AudioClassificationData.from_numpy` ([#1096](https://github.com/PyTorchLightning/lightning-flash/pull/1096)) - - Fixed a bug when using `SpeechRecognitionData.from_files` for training / validating / testing ([#1097](https://github.com/PyTorchLightning/lightning-flash/pull/1097)) - - Fixed a bug when using `SpeechRecognitionData.from_csv` or `from_json` when predicting without targets ([#1097](https://github.com/PyTorchLightning/lightning-flash/pull/1097)) - - Fixed a bug where `SpeechRecognitionData.from_datasets` did not work as expected ([#1097](https://github.com/PyTorchLightning/lightning-flash/pull/1097)) - - Fixed a bug where loading data for prediction with `SemanticSegmentationData.from_folders` raised an error ([#1101](https://github.com/PyTorchLightning/lightning-flash/pull/1101)) - - Fixed a bug when passing a `predict_folder` argument to `from_coco` / `from_voc` / `from_via` in IceVision tasks ([#1102](https://github.com/PyTorchLightning/lightning-flash/pull/1102)) - - Fixed `ObjectDetectionData.from_voc` and `ObjectDetectionData.from_via` ([#1102](https://github.com/PyTorchLightning/lightning-flash/pull/1102)) - - Fixed a bug where `InstanceSegmentationData.from_coco` would raise an error if not using file-based masks ([#1102](https://github.com/PyTorchLightning/lightning-flash/pull/1102)) - - Fixed `InstanceSegmentationData.from_voc` ([#1102](https://github.com/PyTorchLightning/lightning-flash/pull/1102)) - - Fixed a bug when loading tabular data for prediction without a target field / column ([#1114](https://github.com/PyTorchLightning/lightning-flash/pull/1114)) - - Fixed a bug when loading prediction data for graph classification without targets ([#1121](https://github.com/PyTorchLightning/lightning-flash/pull/1121)) - - Fixed a bug where loading Seq2Seq data for prediction would not work if the target field was not present ([#1128](https://github.com/PyTorchLightning/lightning-flash/pull/1128)) - - Fixed a bug where `from_fiftyone` classmethods did not work correctly with a `predict_dataset` ([#1136](https://github.com/PyTorchLightning/lightning-flash/pull/1136)) - - Fixed a bug where the `labels` property would return `None` when using `ObjectDetectionData.from_fiftyone` ([#1136](https://github.com/PyTorchLightning/lightning-flash/pull/1136)) - - Fixed a bug where `TabularData` would not work correctly with no categorical variables ([#1144](https://github.com/PyTorchLightning/lightning-flash/pull/1144)) - - Fixed a bug where loading `TabularForecastingData` for prediction would only yield a single sample per series ([#1149](https://github.com/PyTorchLightning/lightning-flash/pull/1149)) - - Fixed a bug where backbones for the `ObjectDetector`, `KeypointDetector`, and `InstanceSegmentation` tasks were not always frozen correctly when finetuning ([#1163](https://github.com/PyTorchLightning/lightning-flash/pull/1163)) - - Fixed a bug where `DataModule.multi_label` would sometimes be `None` when it had been inferred to be `False` ([#1165](https://github.com/PyTorchLightning/lightning-flash/pull/1165)) ### Removed - Removed the `Seq2SeqData` base class (use `TranslationData` or `SummarizationData` directly) ([#1128](https://github.com/PyTorchLightning/lightning-flash/pull/1128)) - - Removed the ability to attach the `Output` object directly to the model ([#1157](https://github.com/PyTorchLightning/lightning-flash/pull/1157)) ## [0.6.0] - 2021-13-12 diff --git a/flash/__about__.py b/flash/__about__.py index 45560148de..2baba8968c 100644 --- a/flash/__about__.py +++ b/flash/__about__.py @@ -1,18 +1,14 @@ -__version__ = "0.7.0rc0" +__version__ = "0.7.0" __author__ = "PyTorchLightning et al." __author_email__ = "name@pytorchlightning.ai" __license__ = "Apache-2.0" -__copyright__ = f"Copyright (c) 2020-2021, f{__author__}." +__copyright__ = f"Copyright (c) 2020-2022, {__author__}." __homepage__ = "https://github.com/PyTorchLightning/lightning-flash" __docs_url__ = "https://lightning-flash.readthedocs.io/en/stable/" -__docs__ = "Flash is a framework for fast prototyping, finetuning, and solving most standard deep learning challenges" +__docs__ = "Your PyTorch AI Factory - Flash enables you to easily configure and run complex AI recipes." __long_doc__ = """ -Flash is a task-based deep learning framework for flexible deep learning built on PyTorch Lightning. -Tasks can be anything from text classification to object segmentation. -Although PyTorch Lightning provides ultimate flexibility, for common tasks it does not remove 100% of the boilerplate. -Flash is built for applied researchers, beginners, data scientists, Kagglers or anyone starting out with Deep Learning. -But unlike other entry-level frameworks (keras, etc...), Flash users can switch to Lightning trivially when they need -the added flexibility. +Flash makes complex AI recipes for over 15 tasks across 7 data domains accessible to all. +In a nutshell, Flash is the production grade research framework you always dreamed of but didn't have time to build. """ __all__ = [