catalyst -

Published by bagxi over 3 years ago

catalyst - Catalyst 20.12

Published by Scitator almost 4 years ago

[20.12] - 2020-12-20

Added

CVS Logger (#1005)
DrawMasksCallback (#999)
(#1002)
- a few docs
(#998)
- reciprocal_rank metric
- unified recsys metrics preprocessing
(#1018)
- readme examples for all supported metrics under catalyst.metrics
- wrap_metric_fn_with_activation for model outputs wrapping with activation
- extra tests for metrics
(#1039)
- per_class=False option for metrics callbacks
- PrecisionCallack, RecallCallack for multiclass problems
- extra docs

Changed

docs update (#1000)
AMPOptimizerCallback and OptimizerCallback were merged (#1007)
(#1017)
- fixed bug in SchedulerCallback
- Log LRs and momentums for all param groups, not only for the first one
(#1002)
- tensorboard, ipython, matplotlib, pandas, scikit-learn moved to optional requirements
- PerplexityMetricCallback moved to catalyst.callbacks from catalyst.contrib.callbacks
- PerplexityMetricCallback renamed to PerplexityCallback
- catalyst.contrib.utils.confusion_matrix renamed to catalyst.contrib.utils.torch_extra
- many parts of catalyst.data moved to catalyst.contrib.data
- catalyst.data.scripts moved to catalyst.contrib.scripts
- catalyst.utils, catalyst.data.utils and catalyst.contrib.utils restructured
- ReaderSpec renamed to IReader
- SupervisedExperiment renamed to AutoCallbackExperiment
gain functions renamed for dcg/ndcg metrics (#998)
(#1014)
- requirements respecification: catalyst[cv], catalyst[dev], catalyst[log], catalyst[ml], catalyst[nlp],catalyst[tune]
- settings respecification
- extra tests for settings
- contrib refactoring
iou and dice metrics moved to per-class computation (#1031)

Removed

(#1002)
- KNNMetricCallback
- sklearn mode for ConfusionMatrixLogger
- catalyst.data.utils
- unnecessary catalyst.tools.meters
- todos for unnecessary docs
(#1014)
- transformers-based contrib (too unstable)
(#1018)
- ClasswiseIouCallback/ClasswiseJaccardCallback as deprecated on (should be refactored in future releases)

Fixed

prevented modifying config during the experiment and runner initialization (#1004)
a few test for RecSys MAP computation (#1018)
leave batch size the same for default distributed training (#1023)
(#1032)
- Apex: now you can use apex for multiple models training
- Apex: DataParallel is allowed for opt_level other than "O1"

catalyst - Catalyst 20.11

Published by Scitator almost 4 years ago

[20.11] - 2020-11-12

Added

DCG, nDCG metrics (#881)
MAP calculations #968
hitrate calculations [#975] (https://github.com/catalyst-team/catalyst/pull/975)
extra functions for classification metrics (#966)
OneOf and OneOfV2 batch transforms (#951)
precision_recall_fbeta_support metric (#971)
Pruning tutorial (#987)
BatchPrefetchLoaderWrapper (#986)
DynamicBalanceClassSampler (#954)

Changed

update Catalyst version to 20.10.1 for tutorials (#967)
added link to dl-course (#967)
IRunner -> simplified IRunner (#984)
docs were restructured (#985)
set_global_seed moved from utils.seed to utils.misc (#986)

Removed

several deprecated tutorials (#967)
several deprecated func from utils.misc (#986)

Fixed

BatchTransformCallback - add nn.Module transforms support (#951)
moved to contiguous view for accuracy computation (#982)
fixed torch warning on optimizer.py:140 (#979)

catalyst - Catalyst 20.10.1

Published by Scitator almost 4 years ago

[20.10.1] - 2020-10-15

Added

MRR metrics calculation (#886)
docs for MetricCallbacks (#947)
SoftMax, CosFace, ArcFace layers to contrib (#939)
ArcMargin layer to contrib (#957)
AdaCos to contrib (#958)
Manual SWA to utils (#945)

Changed

fixed path to CHANGELOG.md file and add information about unit test to PULL_REQUEST_TEMPLATE.md ([#955])(https://github.com/catalyst-team/catalyst/pull/955)
catalyst-dl tune config specification - now optuna params are grouped under study_params (#947)
IRunner._prepare_for_stage logic moved to IStageBasedRunner.prepare_for_stage (#947)
- now we create components in the following order: datasets/loaders, model, criterion, optimizer, scheduler, callbacks
MnistMLDataset and MnistQGDataset data split logic - now targets of the datasets are disjoint (#949)
architecture redesign (#953)
- experiments, runners, callbacks grouped by primitives under catalyst.experiments/catalyst.runners/catalyst.callbacks respectively
- settings and typing moved from catalyst.tools.* to catalyst.*
- utils moved from catalyst.*.utils to catalyst.utils
swa moved to catalyst.utils (#963)

Removed

Fixed

AMPOptimizerCallback - fix grad clip fn support (#948)
removed deprecated docs types (#947) (#952)
docs for a few files (#952)
extra backward compatibility fixes (#963)

catalyst - Catalyst 20.09.1

Published by Scitator almost 4 years ago

[20.09.1] - 2020-09-25

Added

Runner registry support for Config API (#936)
catalyst-dl tune command - Optuna with Config API integration for AutoML hyperparameters optimization (#937)
OptunaPruningCallback alias for OptunaCallback (#937)
AdamP and SGDP to catalyst.contrib.nn.criterion (#942)

Changed

Config API components preparation logic moved to utils.prepare_config_api_components (#936)

Removed

Fixed

Logging double logging :) (#936)
CMCCallback (#941)

catalyst - Catalyst 20.09

Published by Scitator almost 4 years ago

[20.09] - 2020-09-07

Added

MovieLens dataset loader (#903)
force and bert-level keywords to catalyst-data text2embedding (#917)
OptunaCallback to catalyst.contrib (#915)
DynamicQuantizationCallback and catalyst-dl quantize script for fast quantization of your model (#890)
Multi-scheduler support for multi-optimizer case (#923)
Native mixed-precision training support (#740)
OptiomizerCallback - flag use_fast_zero_grad for faster (and hacky) version of optimizer.zero_grad() (#927)
IOptiomizerCallback, ISchedulerCallback, ICheckpointCallback, ILoggerCallback as core abstractions for Callbacks (#933)
flag USE_AMP for PyTorch AMP usage (#933)

Changed

Pruning moved to catalyst.dl (#933)
default USE_APEX changed to 0 (#933)

Removed

Fixed

autoresume option for Config API (#907)
a few issues with TF projector (#917)
batch sampler speed issue (#921)
add apex key-value optimizer support (#924)
runtime warning for PyTorch 1.6 (920)
Apex synbn usage (920)
Catalyst dependency on system git (922)

catalyst - Catalyst 20.08

Published by Scitator almost 4 years ago

[20.08] - 2020-08-09

Added

CMCScoreCallback (#880)
kornia augmentations BatchTransformCallback (#862)
average_precision and mean_average_precision metrics (#883)
MultiLabelAccuracyCallback, AveragePrecisionCallback and MeanAveragePrecisionCallback callbacks (#883)
minimal examples for multiclass and multilabel classification (#883)
experimental TPU support (#893)
add Imagenette, Imagewoof, and Imagewang datasets (#902)
IMetricCallback, IBatchMetricCallback, ILoaderMetricCallback, BatchMetricCallback, LoaderMetricCallback abstractions (#897)
HardClusterSampler inbatch sampler (#888)

Changed

all registries merged to one catalyst.registry (#883)
mean_average_precision logic merged with average_precision (#897)
all imports moved to absolute (#905)
catalyst.contrib.data merged to catalyst.data (#905)
{breaking} Catalyst transform ToTensor was renamed to ImageToTensor (#905)
TracerCallback moved to catalyst.dl (#905)
ControlFlowCallback, PeriodicLoaderCallback moved to catalyst.core (#905)

Removed

average_accuracy and mean_average_accuracy metrics (#883)
MultiMetricCallback abstraction (#897)

Fixed

utils.tokenize_text typo with punctuation (#880)
ControlFlowCallback logic (#892)
docs (#897)

catalyst - Catalyst 20.07

Published by Scitator almost 4 years ago

[20.07] - 2020-07-06

Added

log parameter to WandbLogger (#836)
hparams experiment property (#839)
add docs build on push to master branch (#844)
WrapperCallback and ControlFlowCallback (#842)
BatchOverfitCallback (#869)
overfit flag for Config API (#869)
InBatchSamplers: AllTripletsSampler and HardTripletsSampler (#825)

Changed

Renaming (#837)
- SqueezeAndExcitation -> cSE
- ChannelSqueezeAndSpatialExcitation -> sSE
- ConcurrentSpatialAndChannelSqueezeAndChannelExcitation -> scSE
- _MetricCallback -> IMetricCallback
- dl.Experiment.process_loaders -> dl.Experiment._get_loaders
LRUpdater become abstract class (#837)
calculate_confusion_matrix_from_arrays changed params order (#837)
dl.Runner.predict_loader uses _prepare_inner_state and cleans experiment (#863)
toml to the dependencies (#872)

Removed

crc32c dependency (#872)

Fixed

workflows/deploy_push.yml failed to push some refs (#864)
.dependabot/config.yml contained invalid details (#781)
LanguageModelingDataset (#841)
global_* counters in Runner (#858)
EarlyStoppingCallback considers first epoch as bad (#854)
annoying numpy warning (#860)
PeriodicLoaderCallback overwrites best state (#867)
OneCycleLRWithWarmup (#851)

catalyst - Catalyst 20.06

Published by Scitator over 4 years ago

[20.06] - 2020-06-04

Added

Mergify (#831)
PerplexityMetricCallback (#819)
PeriodicLoaderRunnerCallback (#818)

Changed

docs structure were updated during (#822)
utils.process_components moved from utils.distributed to utils.components (#822)
catalyst.core.state.State merged to catalyst.core.runner._Runner (#823) (backward compatibility included)
- catalyst.core.callback.Callback now works directly with catalyst.core.runner._Runner
- state_kwargs renamed to stage_kwargs

Removed

Fixed

added missed dashes in docker perfixes (#828)

[20.05.1] - 2020-05-23

Added

Circle loss implementation (#802)
BatchBalanceSampler for metric learning and classification (#806)
CheckpointCallback: new argument load_on_stage_start which accepts str and Dict[str, str] (#797)
LanguageModelingDataset to catalyst[nlp] (#808)
Extra counters for batches, loaders and epochs (#809)
TracerCallback (#789)

Changed

CheckpointCallback: additional logic for argument load_on_stage_end - accepts str and Dict[str, str] (#797)
counters names for batches, loaders and epochs (#809)
utils.trace_model: changed logic - runner argument was changed to predict_fn (#789)
redesigned contrib.data and contrib.datasets (#820)
catalyst.utils.meters moved to catalyst.tools (#820)
catalyst.contrib.utils.tools.tensorboard moved to catalyst.contrib.tools (#820)

Removed

Fixed

device selection fix for #798 (#815)
batch size counting fix for #799 and #755 issues (#809)

catalyst - Catalyst 20.05.1

Published by Scitator over 4 years ago

catalyst - Catalyst 20.06rc1

Published by Scitator over 4 years ago

catalyst - Catalyst 20.05

Published by Scitator over 4 years ago

catalyst - Catalyst 20.03.1

Published by Scitator over 4 years ago

tl;dr

We finally organise Experiment-Runner-State-Callback as it should be.
We also have great documentation update!

Core

Experiment - an abstraction that contains information about the experiment – a model, a criterion, an optimizer, a scheduler, and their hyperparameters. It also contains information about the data and transformations used. In general, the Experiment knows what you would like to run.
Runner - a class that knows how to run an experiment. It contains all the logic of how to run the experiment, stages, epoch and batches.
State - some intermediate storage between Experiment and Runner that saves the current state of the Experiments – model, criterion, optimizer, schedulers, metrics, loaders, callbacks, etc

Callback - a powerful abstraction that lets you customize your experiment run logic. To give users maximum flexibility and extensibility we allow callback execution anywhere in the training loop

on_stage_start
    on_epoch_start
       on_loader_start
           on_batch_start
           # ... 
       on_batch_end
    on_epoch_end
on_stage_end

on_exception

How to combine them together?

First of all - just read the docs for State, Experiment, Runner and Callback abstractions.
Long story short, State just saves everything during experiment and passes to every Callback in Experiment through Runner.run_event.
For example, usual case, for some custom metric implementation, all you need to do is

from catalyst.dl import Callback, State

class MyPureMetric(Callback):
  def on_batch_end(self, state: State):
    """To store batch-based metrics"""
    state.batch_metrics[{metric_name}] = metric_value

  def on_loader_end(self, state: State):
  	"""To store loader-based metrics"""
    state.loader_metrics[{metric_name}] = metric_value

  def on_epoch_end(self, state: State):
    """To store epoch-based metrics"""
    state.epoch_metrics[{metric_name}] = metric_value

There are coming many more Catalyst concepts, tutorials and docs in near future.

Callback updates

CheckRunCallback - allows you to check the pipeline correctness during long-run training
TimerCallback - enables / disables the calculation of the work speed of your training, like data time per batch, model time per batch, samples per second
MetricManagerCallback - transfers torch tensors to numpy and calculates statistics
ValidationManagerCallback - collects validation metrics and checks if it's the best epoch

Catalyst best practices

Working with Catalyst.DL it's better to import everything in straightforward way like

from catalyst.dl import SomethingGreat
from catalyst.dl import utils

utils.do_something_cool()

Breaking changes

CriterionAggregatorCallback moved to catalyst.contrib and will be deprecated in 20.04 release.
For SchedulerCallback reduce_metric was renamed to reduced_metric :)
We have update metric recording mechanism for State-Callback,

Future work

During 20.03 -> 20.04 releases, we are going to deprecate all SomeContribRunner and transfer them to SomeContribLogger as more general purpose solution.

catalyst - 20.02.2

Published by Scitator over 4 years ago

Catalyst 20.02.2

Overall

scripts update
formatting update
project manifesto https://github.com/catalyst-team/catalyst/pull/532
Alchemy intro https://github.com/catalyst-team/catalyst/pull/563 https://github.com/catalyst-team/catalyst/pull/564
Travis -> TeamCity CI https://github.com/catalyst-team/catalyst/pull/597 https://github.com/catalyst-team/catalyst/pull/616 https://github.com/catalyst-team/catalyst/pull/634
Core init https://github.com/catalyst-team/catalyst/pull/578
tests global update https://github.com/catalyst-team/catalyst/pull/661

DL

config wizard https://github.com/catalyst-team/catalyst/pull/499
Triplet loss contribution https://github.com/catalyst-team/catalyst/pull/504
TelegramLogger support https://github.com/catalyst-team/catalyst/pull/471
Trace support for fp16 and Nvidia Apex https://github.com/catalyst-team/catalyst/pull/497
runner params fix https://github.com/catalyst-team/catalyst/pull/545 https://github.com/catalyst-team/catalyst/pull/633
weighted sum support for loss calculation https://github.com/catalyst-team/catalyst/pull/535
classwise IOU support https://github.com/catalyst-team/catalyst/pull/533
BERT text classification example https://github.com/catalyst-team/catalyst/pull/540
Ralamb contribution https://github.com/catalyst-team/catalyst/pull/551
samplers params support https://github.com/catalyst-team/catalyst/pull/550
KNNMetricCallback contribution https://github.com/catalyst-team/catalyst/pull/560
MetricAggregationCallback contribution https://github.com/catalyst-team/catalyst/pull/591
text to embeddings script https://github.com/catalyst-team/catalyst/pull/596 https://github.com/catalyst-team/catalyst/pull/601
Several metric learning features https://github.com/catalyst-team/catalyst/pull/589 https://github.com/catalyst-team/catalyst/pull/598 https://github.com/catalyst-team/catalyst/pull/599
Neptune integration https://github.com/catalyst-team/catalyst/pull/571
transforms params support https://github.com/catalyst-team/catalyst/pull/595 https://github.com/catalyst-team/catalyst/pull/604
SMP integration https://github.com/catalyst-team/catalyst/pull/600
GAN 2.0 https://github.com/catalyst-team/catalyst/pull/607 https://github.com/catalyst-team/catalyst/pull/585
Cutmix callback contribution https://github.com/catalyst-team/catalyst/pull/635
RMS Normalization contribution https://github.com/catalyst-team/catalyst/pull/649
EarlyStoppingCallback fix https://github.com/catalyst-team/catalyst/pull/664
better distributed & slurm support https://github.com/catalyst-team/catalyst/pull/639 https://github.com/catalyst-team/catalyst/pull/629 https://github.com/catalyst-team/catalyst/pull/662 https://github.com/catalyst-team/catalyst/pull/628

catalyst - Catalyst 19.11

Published by Scitator almost 5 years ago

Catalyst 19.10 -> 19.11

Achievements

NeurIPS 2019: Learn to Move - Walk Around, 2nd place

catalyst - Catalyst 19.10

Published by Scitator about 5 years ago

Catalyst 19.09 -> 19.10

Achievements

NeurIPS 2019: Recursion Cellular Image Classification

4th place solution writeup
8th place solution

Kaggle integration

Weights & Biases integration

Overall

LAMA refactoring https://github.com/catalyst-team/catalyst/pull/353
TemporalConcatPooling https://github.com/catalyst-team/catalyst/pull/355
Extensions on/off support https://github.com/catalyst-team/catalyst/pull/368
Hall of Glory update https://github.com/catalyst-team/catalyst/pull/385
binary_mask_to_overlay_image refactoring https://github.com/catalyst-team/catalyst/pull/387
get_utcnow_time feature https://github.com/catalyst-team/catalyst/pull/389
image preprocessing refactoring https://github.com/catalyst-team/catalyst/pull/398
TF seed fix https://github.com/catalyst-team/catalyst/pull/399
shorthand for bool flag https://github.com/catalyst-team/catalyst/pull/405

DL

Table data tutorial https://github.com/catalyst-team/catalyst/pull/351
Wandb integration tutorial https://github.com/catalyst-team/catalyst/commit/365ed7d6fdb9e5aa515ef52612e403fdd3270ac8
lovasz loss fix https://github.com/catalyst-team/catalyst/pull/359
SupervisedRunner, one loader handling https://github.com/catalyst-team/catalyst/pull/362
SupervisedRunner.predict_loader, model support https://github.com/catalyst-team/catalyst/pull/367
resume from previous stage fix https://github.com/catalyst-team/catalyst/pull/328
CriterionAggregator https://github.com/catalyst-team/catalyst/pull/361
Weight decay decoupling https://github.com/catalyst-team/catalyst/pull/370
Binary class support for AccuracyCallback https://github.com/catalyst-team/catalyst/pull/374
SupervisedRunner.train, resume feature https://github.com/catalyst-team/catalyst/pull/377
Apex distributed synchronised batchnorm support https://github.com/catalyst-team/catalyst/pull/379
Catalyst Init feature https://github.com/catalyst-team/catalyst/pull/372
Dict transform support for MergeDataset https://github.com/catalyst-team/catalyst/pull/388
BCEDiceLoss weights support https://github.com/catalyst-team/catalyst/pull/394
Multi-label accuracy support https://github.com/catalyst-team/catalyst/pull/381
PyTorch loaders additional params support https://github.com/catalyst-team/catalyst/pull/397

RL

RL networks refactoring – MLP and CNN policies support https://github.com/catalyst-team/catalyst/pull/352
Mongo reconnection support https://github.com/catalyst-team/catalyst/pull/354
RL tests refactoring https://github.com/catalyst-team/catalyst/pull/357 https://github.com/catalyst-team/catalyst/pull/369
EnvWrapper feature https://github.com/catalyst-team/catalyst/pull/358
Raw reward fix https://github.com/catalyst-team/catalyst/pull/373
deterministic RL https://github.com/catalyst-team/catalyst/pull/393
Large files support for MongoDB https://github.com/catalyst-team/catalyst/pull/396

catalyst - Catalyst 19.09

Published by Scitator about 5 years ago

Catalyst 19.08 -> 19.09

Ecosystem

We are happy to announce MLComp release – a distributed DAG (Directed acyclic graph) framework for machine learning with UI. Powered by Catalyst.Team.

We also release a detailed classification tutorial and comprehensive classification pipeline.