3.0.4

What's Changed

• Reorganisation README by @Shani-Perl in #526
• Apply black formatting on pretrained model zoo by @BloodAxe in #524
• Remove imports from factory.init by @Louis-Dupont in #516
• Feature/sg 416 albumentations plugin for classification by @shaydeci in #495
• Feature/sg 404 ssd reproduce by @BloodAxe in #525
• Added indexing support for meshgrid by @BloodAxe in #528
• Feature/al 441 ptq detection by @spsancti in #527
• Add DDP doc by @Louis-Dupont in #523
• SG-448- dataset_params and arch_params logging by @shaydeci in #532
• Fix clearml comment by @Louis-Dupont in #530
• Bug/sg 000 torch version fix by @shaydeci in #535
• fix pplite-seg prep conversion by @lkdci in #538
• SGLogger fix - add warning and multiprocess_safe system_monitoring by @Louis-Dupont in #534
• Add self to recipes by @Louis-Dupont in #539
• replace head for UNet module by @lkdci in #543
• Rename gpu_mode to multi_gpu and setup_gpu_mode to setup_device by @Louis-Dupont in #517
• Add version_base to hydra to remove warning by @Louis-Dupont in #541
• Hotfix/sg 000 add data structure and link to cityscape desc by @Louis-Dupont in #550
• add stringcase by @ofrimasad in #553
• Hotfix/sg 000 add data structure to cityscape desc v2 by @Louis-Dupont in #551
• Improve PascalVOC detection error msg by @Louis-Dupont in #556
• Improve evaluate_from_recipe usability by @Louis-Dupont in #558
• Feature/sg 458 support for any user dataloader using dataset registry by @shaydeci in #547
• Features/sg 409 check all params used by @BloodAxe in #546
• Split and rename the modules from super_gradients.common.environment by @Louis-Dupont in #548
• Feature/sg 468 detection transform support for any number of channels by @shaydeci in #559
• Feature/sg 518 update evaluate from recipe to add output by @Louis-Dupont in #560
• error raised for torch version and formatting by @shaydeci in #540
• Feature/sg 356 ddp silent mode and multi process safe docs by @shaydeci in #563
• Apply black on some files by @Louis-Dupont in #565
• Fix runtime warning of accessing non-contiguous tensor in mAP metric by @BloodAxe in #566
• fix F1 Precision Recall to represent all the defined rage and not onl… by @ofrimasad in #564
• Fix Multigpu.OFF factory by @Louis-Dupont in #569
• Import AutoLoggerConfig and ConsoleSink separatly by @Louis-Dupont in #567
• remove user guide from docs by @ofrimasad in #570
• new generated docs by @ofrimasad in #572
• update documentation by @ofrimasad in #573
• Make ConvBNReLU a subclass of ConvBNAct to keep backward compatibililty by @BloodAxe in #554
• Update README.md by @ofrimasad in #575
• Apply black on losses by @Louis-Dupont in #576
• Feature/sg 516 support head replacement for local pretrained weights unknown dataset by @shaydeci in #578
• Fix bug of not respecting dilation argument in RepVGGBlock by @BloodAxe in #555
• Feature/sg 431 check classes by @BloodAxe in #531
• register unet cls module by @lkdci in #580
• feature: Accept an arbitrary WandB ID by @yurkovak in #582
• Bug/sg 512 shuffle bugfix in recipe datalaoders by @shaydeci in #581
• Added import of quantized modules so they get registered by @spsancti in #542
• Use unpack_batch_items on first batch (Respect additional batch items) by @BloodAxe in #584

Full Changelog: 3.0.3...3.0.4

3.0.3

• ClearML integration.
• Console logging + upload (Lab visibility).
• Epoch summary (visual improvement).
• Registry for Phase Callbacks and Transforms.
• Infrastructure for crash tips.
• Detection output adapter.
• Train from recipe example- external dataset.
• QAT infrastructure + examples.
• System Logger (i.e device attributes etc).

3.0.2

• dataloaders.get() supports Dataset objects.
• Behind the scenes sampler handling (warning + initialization).
• Black formatter.
• New DetectionBase.
• Logs moved from ~/sg_log/{module_name}
• CSP Resnet Backbone, new place for modules in repo.
• Plenty new factories.
• Message on effective batch size
• External checkpoints resume- fix
• Passing external classes list for CocoDetection
• Fix for loading weights from the platform

3.0.1

What's new?

• Eval recipe- perform validation by recipe name.
• Supported strings class for convenient autocomplete in IDE's.
• Registry: metrics, dataloaders, models, losses (so they can be passed as strings when using train_from_config).
• Return model, results from train
• Load backbone fix https://github.com/Deci-AI/super-gradients/pull/408/files
• PPLiteSeg training recipes.
• AWS env check removal
• New resolvers support: "+" "if" for yaml recipes (i.e Hydra resolvers).
• Support for custom STDC.
• ShelfNet "classes_num" -> "num_classes" bug fix.
• Improve cross-platform compatibility when parsing a readme description.
• Added the ability to download and import external code for models from ADK.

3.0.0

• DatasetInterface class removal- refactored as torch.DataLoader objects configured by src/training/recipes, using super_gradients.dataloaders.get() (see new updated tutorials and snippets).

• Trainer.build_model() removal- models initialisation refactored with super_gradients.models.get() (see updated tutorials and notebooks).

• Coded DDP launch (no need for python -m torch.distributed.launch ...), see new snippets here .

• Updated notebooks, tutorials and code snippets in readme.md.

• Extract recipes training hyper_params config with super_gradients.training_hyperparams.get() (see updated tutorials and notebooks).

• Simplfied resume- now passed through train_params in SgTrainer.train() (see updated snippets in readme.md).

• Removal of "loss_loggging_items_names" from train_params in Trainer.train().

• Trainer.init old, unnecessary args removed.

• Add support for getting models from Deci's platform using super_gradients.models.get(), more info regarding Deci's platform in readme.md.

2.6.0: Launch DDP with script (#350)

This GitHub Release was done automatically by CircleCI

2.5.0: Feature/sg 216 remove dataset interface (#356)

This GitHub Release was done automatically by CircleCI

2.2.0: Hotfix detection cache: Add hash in name when caching (#308)

This GitHub Release was done automatically by CircleCI

2.1.0

• YoloX architectures.
• SSDLite Mobilenet V2 COCO recipe
• QAT support with Nvidia's pytorch-quantization (optional dependency).
• COCO mAP calculation support in DDP (torch metric object, supports "crowd" labels).
• Pre_prediction_callback- support for input and targets manipulation right before forward pass + multiscaling pre_prediction_callbacks that work out of the box in DDP (classification and Object detection).
• Training stage switch callback to support multi-stage training.

2.0.1

This GitHub Release was done automatically by CircleCI