KevinMusgrave/powerful-benchmarker issues and pull requests

#114 - Add code highlighting in the README

Pull Request - State: closed - Opened by bryant1410 9 months ago - 1 comment

#113 - Get error when execute script in 'powerful-benchmarker' directory.

Issue - State: open - Opened by hwopark over 1 year ago - 1 comment

#112 - Dev2

Pull Request - State: closed - Opened by KevinMusgrave over 1 year ago

#111 - updated resilience plot

Pull Request - State: closed - Opened by KevinMusgrave over 1 year ago

#110 - epoch0

Pull Request - State: closed - Opened by KevinMusgrave over 1 year ago

#109 - Default split manager

Issue - State: closed - Opened by tfrsw over 1 year ago - 4 comments
Labels: question, metric-learning

#108 - Bump torch from 1.10.1+cu111 to 1.13.1

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#107 - dev into master

Pull Request - State: closed - Opened by KevinMusgrave over 1 year ago

#106 - Dev into master

Pull Request - State: closed - Opened by KevinMusgrave over 1 year ago

#105 - dev into master

Pull Request - State: closed - Opened by KevinMusgrave almost 2 years ago

#104 - dev into master

Pull Request - State: closed - Opened by KevinMusgrave about 2 years ago

#103 - Use --extend-ignore in linter rule

Issue - State: closed - Opened by KevinMusgrave about 2 years ago
Labels: domain-adaptation

#102 - Updated the links of Cars196 dataset

Pull Request - State: closed - Opened by ajktym94 over 2 years ago - 1 comment

#101 - validator_tests/delete_pkls: Change --validator flag to --prefix

Issue - State: open - Opened by KevinMusgrave over 2 years ago
Labels: domain-adaptation

#100 - Dev into master

Pull Request - State: closed - Opened by KevinMusgrave over 2 years ago

#99 - More readme updates

Pull Request - State: closed - Opened by KevinMusgrave over 2 years ago

#98 - Added readme to validator_tests

Pull Request - State: closed - Opened by KevinMusgrave over 2 years ago

#97 - Merge dev into master

Pull Request - State: closed - Opened by KevinMusgrave over 2 years ago

#96 - How to evaluate an experiment (domain adaptation)

Issue - State: closed - Opened by pablopalafox over 2 years ago - 2 comments
Labels: bug, domain-adaptation

#95 - Pretraining on source

Issue - State: closed - Opened by pablopalafox over 2 years ago - 3 comments
Labels: domain-adaptation

#94 - Trouble Reproducing ArcFace Results

Issue - State: closed - Opened by mdemoor226 over 2 years ago - 5 comments
Labels: metric-learning

#93 - Uda prep

Pull Request - State: closed - Opened by KevinMusgrave over 2 years ago

#91 - Compatibility with latest pytorch-metric-learning library

Issue - State: closed - Opened by aidinbii about 3 years ago - 1 comment
Labels: bug, metric-learning

#90 - Proxy NCA And Softmax Scale / Label Smoothing

Issue - State: closed - Opened by lou-k about 3 years ago - 3 comments
Labels: metric-learning

#89 - The ~INT_BAYESIAN~ flag does not seem to work

Issue - State: closed - Opened by cerebrai about 3 years ago - 1 comment
Labels: bug, metric-learning

#88 - Reproducing benchmark results only gives validation accuracy

Issue - State: closed - Opened by cerebrai about 3 years ago - 2 comments
Labels: metric-learning

#87 - Question on SoftTripleLoss Bayes hyperparameter tuning

Issue - State: closed - Opened by lindsey98 about 3 years ago - 2 comments
Labels: metric-learning

#86 - Clarification on precision_at_k computation

Issue - State: closed - Opened by lindsey98 about 3 years ago - 2 comments
Labels: metric-learning

#85 - Set accuracy evaluator to only get precision_at_k=1

Issue - State: closed - Opened by lindsey98 about 3 years ago - 1 comment
Labels: metric-learning

#84 - Test on benchmark Error

Issue - State: closed - Opened by alpoler over 3 years ago - 1 comment
Labels: metric-learning

#83 - Make trained models available on torch.hub

Issue - State: closed - Opened by KevinMusgrave over 3 years ago
Labels: enhancement, metric-learning

#82 - Cross Batch Memory error

Issue - State: closed - Opened by atkvb over 3 years ago - 4 comments
Labels: bug, metric-learning

#81 - Stanford Online Product training scheme

Issue - State: closed - Opened by elias-ramzi over 3 years ago - 5 comments
Labels: metric-learning

#80 - Can I add my own learning rate scheduler during training?

Issue - State: closed - Opened by weikunhan over 3 years ago - 3 comments
Labels: question, metric-learning

#79 - Whem MClassPerSampler is equal to 1, Error occurs

Issue - State: closed - Opened by alpoler over 3 years ago - 1 comment
Labels: bug, metric-learning

#78 - Bug due to ReLU(inplace=True) with TrainWithClassifier

Issue - State: closed - Opened by alpoler over 3 years ago - 3 comments
Labels: bug, metric-learning

#77 - Error when run experiment reproducing

Issue - State: closed - Opened by liuhl-source almost 4 years ago - 7 comments
Labels: bug, metric-learning

#76 - Customized loss function need to know the train/val split

Issue - State: closed - Opened by ouyangzhibo almost 4 years ago - 3 comments
Labels: frequently asked questions, question, metric-learning

#75 - Validation label appears in a training batch

Issue - State: closed - Opened by btseytlin almost 4 years ago - 2 comments
Labels: frequently asked questions, question, metric-learning

#74 - Fix number of epoch train for all splits

Issue - State: closed - Opened by alpoler almost 4 years ago - 1 comment
Labels: frequently asked questions, question, metric-learning

#73 - Make it easier to use predefined split manager

Issue - State: closed - Opened by KevinMusgrave almost 4 years ago
Labels: enhancement, metric-learning

#72 - Specifying optimizer scheduler and gradient clipper is broken

Issue - State: closed - Opened by KevinMusgrave almost 4 years ago
Labels: bug, metric-learning

#71 - Resume, but do reproduction only

Issue - State: closed - Opened by btseytlin almost 4 years ago - 3 comments
Labels: frequently asked questions, question, metric-learning

#70 - ProxyNCA Loss optimizer step location

Issue - State: closed - Opened by alpoler almost 4 years ago - 3 comments
Labels: frequently asked questions, question, metric-learning

#69 - Save and evaluate final epoch, for cases where num_epochs_train is not a multiple of save_interval

Issue - State: closed - Opened by KevinMusgrave almost 4 years ago
Labels: enhancement, metric-learning

#68 - Reduce dependency on dataset transform attribute

Issue - State: closed - Opened by KevinMusgrave about 4 years ago
Labels: enhancement, metric-learning

#67 - Remove 3 lines in base api parser that depends on labels in dataset

Issue - State: closed - Opened by KevinMusgrave about 4 years ago
Labels: enhancement, metric-learning

#66 - How to select which GPU to use?

Issue - State: closed - Opened by btseytlin about 4 years ago - 2 comments
Labels: frequently asked questions, question, metric-learning

#56 - Add wandb as an option for logging

Issue - State: closed - Opened by KevinMusgrave about 4 years ago
Labels: enhancement, metric-learning

#54 - Update loss function factory to be compatible with pytorch-metric-learning v0.9.90

Issue - State: closed - Opened by KevinMusgrave about 4 years ago
Labels: enhancement, fixed in dev version, metric-learning

#53 - Add ePillID to default datasets

Issue - State: closed - Opened by KevinMusgrave about 4 years ago
Labels: enhancement, metric-learning

#52 - Reduce Dependency on Command Line

Issue - State: closed - Opened by Queuecumber about 4 years ago
Labels: enhancement, metric-learning

#51 - How to use my own dataset for benchmarking metric learning?

Issue - State: closed - Opened by dorcus01 about 4 years ago - 2 comments
Labels: frequently asked questions, question, metric-learning

#50 - How to use my own dataset with powerful-benchmarker?

Issue - State: closed - Opened by dorcus01 about 4 years ago
Labels: metric-learning

#49 - Model 0 should be saved as "best" before training starts

Issue - State: closed - Opened by KevinMusgrave about 4 years ago
Labels: bug, fixed in dev version, metric-learning

#48 - Use benchmark for new datasets

Issue - State: closed - Opened by gustmd0121 about 4 years ago - 10 comments
Labels: frequently asked questions, question, metric-learning

#47 - Default to BaseAPIParser if a matching API parser is not found

Issue - State: closed - Opened by KevinMusgrave about 4 years ago
Labels: enhancement, fixed in dev version, metric-learning

#45 - Make sure stuff works before ECCV

Issue - State: closed - Opened by KevinMusgrave about 4 years ago
Labels: metric-learning

#44 - where's the code corresponding to "special_split_scheme"?

Issue - State: closed - Opened by davidsonic about 4 years ago - 4 comments
Labels: documentation, question, metric-learning

#43 - Try to modularize the meta testers

Issue - State: closed - Opened by KevinMusgrave about 4 years ago - 1 comment
Labels: enhancement, metric-learning

#42 - Move trainer and tester config options into object form

Issue - State: closed - Opened by KevinMusgrave about 4 years ago - 1 comment
Labels: enhancement, metric-learning

#41 - Specify some GPUs in parallel

Issue - State: closed - Opened by CaiSudao over 4 years ago - 2 comments
Labels: question, metric-learning

#40 - Split dataset

Issue - State: closed - Opened by CaiSudao over 4 years ago - 9 comments
Labels: frequently asked questions, question, metric-learning

#39 - Recreate Test set

Issue - State: closed - Opened by marijnl over 4 years ago - 2 comments
Labels: frequently asked questions, question, metric-learning

#38 - Reason to freeze_batchnorm

Issue - State: closed - Opened by marijnl over 4 years ago - 1 comment
Labels: frequently asked questions, question, metric-learning

#37 - How to use the Bayesian optimization to tune ArcFaceLoss hyperparameters?

Issue - State: closed - Opened by CaiSudao over 4 years ago - 2 comments
Labels: frequently asked questions, question, metric-learning

#36 - The condition for randomize_embedder should be reversed

Issue - State: closed - Opened by KevinMusgrave over 4 years ago - 1 comment
Labels: bug, fixed in dev version, metric-learning

#35 - Subclass BaseAPIParser has no attribute 'split_manager'

Issue - State: closed - Opened by marijnl over 4 years ago - 1 comment
Labels: metric-learning

#34 - Reproducing benchmark results

Issue - State: closed - Opened by CaiSudao over 4 years ago - 5 comments
Labels: frequently asked questions, metric-learning

#33 - Bayesian optimization to tune hyperparameters coding problem.

Issue - State: closed - Opened by CaiSudao over 4 years ago - 17 comments
Labels: frequently asked questions, metric-learning

#32 - Test and validation accuracies

Issue - State: closed - Opened by marijnl over 4 years ago - 6 comments
Labels: question, metric-learning

#31 - Run multiple bayes trails in parallel

Issue - State: closed - Opened by marijnl over 4 years ago - 2 comments
Labels: enhancement, metric-learning

#30 - Allow visualizer to be specified

Issue - State: closed - Opened by KevinMusgrave over 4 years ago - 5 comments
Labels: enhancement, fixed in dev version, metric-learning

#29 - Remove hard-coded value in meta_eval

Issue - State: closed - Opened by KevinMusgrave over 4 years ago - 1 comment
Labels: enhancement, fixed in dev version, metric-learning

#28 - Bayesian optimization with custom dataset

Issue - State: closed - Opened by marijnl over 4 years ago - 3 comments
Labels: bug, metric-learning

#27 - BayesOptRunner should only launch SingleExperimentRunner when necessary

Issue - State: closed - Opened by KevinMusgrave over 4 years ago
Labels: enhancement, metric-learning

#26 - best_epoch should be determined from filename

Issue - State: closed - Opened by KevinMusgrave over 4 years ago - 1 comment
Labels: enhancement, fixed in dev version, metric-learning