CNNTrainLang merge requestshttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests2021-02-09T15:48:56+01:00https://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/34bug fixes2021-02-09T15:48:56+01:00Sascha Dewesbug fixesEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/33Weight Initializer2021-02-03T18:49:02+01:00Sascha DewesWeight InitializerAdded weight initializer symbolsAdded weight initializer symbolsEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/32Execution semantics2021-01-18T14:06:18+01:00Malte HeithoffExecution semanticsCommon VersioningCommon VersioningEvgeny KusmenkoMalte HeithoffEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/31merge Master into dev2020-10-26T16:01:27+01:00Evgeny Kusmenkomerge Master into devhttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/30Develop2020-10-26T16:06:54+01:00Julian Johannes Steinsberger-DührßenDevelopEvgeny KusmenkoJulian Johannes Steinsberger-DührßenEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/29Develop2020-08-27T01:19:56+02:00Julian Johannes Steinsberger-DührßenDevelopEvgeny KusmenkoJulian Johannes Steinsberger-DührßenEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/28Add optional package declaration to CNNTrain grammar && update version number2020-06-29T22:35:19+02:00Han ZhangAdd optional package declaration to CNNTrain grammar && update version numberAdd optional package declaration into CNNTrain grammar and update the project version numberAdd optional package declaration into CNNTrain grammar and update the project version numberEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/27Develop2020-04-08T11:32:35+02:00Julian DierkesDevelopExtendend language by features for GAN training configurationsExtendend language by features for GAN training configurationsEvgeny KusmenkoJulian DierkesEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/26Added several parameters2020-01-16T16:19:08+01:00Sebastian NickelsAdded several parameters@sebastian.nickels:
Added checkpoint_period, log_period
Added eval_train
Added shuffle_data
Added clip_global_grad_norm
@dierkes-julian:
Added constraint losses@sebastian.nickels:
Added checkpoint_period, log_period
Added eval_train
Added shuffle_data
Added clip_global_grad_norm
@dierkes-julian:
Added constraint lossesEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/25Added attention-related params2019-12-02T12:27:20+01:00Christian FußAdded attention-related params* Added SoftmaxCrossEntropyIgnoreIndices loss, and the "ignore_indices" parameter
* Added "saveAttentionImage" and "useTeacherForcing" flags* Added SoftmaxCrossEntropyIgnoreIndices loss, and the "ignore_indices" parameter
* Added "saveAttentionImage" and "useTeacherForcing" flagsEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/24update the version number of common-monticar2019-11-12T20:54:12+01:00Yuyuan Liuupdate the version number of common-monticar@kusmenko I have updated the version number of common-monticar in pom.xml.@kusmenko I have updated the version number of common-monticar in pom.xml.Evgeny KusmenkoYuyuan LiuEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/23Tensorflow group2019-10-18T12:48:13+02:00Evgeny KusmenkoTensorflow grouphttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/22Cocos fix2019-10-07T15:26:56+02:00Nicola GattoCocos fix- Increment version to 0.3.7-SNAPSHOT
- Synchronize develop with master branch and resolve merge conflicts
- Fix Cocos that checks if OU parameter fits to trained architecture
- Fix RL parameter check
- Split dimension check of actions i...- Increment version to 0.3.7-SNAPSHOT
- Synchronize develop with master branch and resolve merge conflicts
- Fix Cocos that checks if OU parameter fits to trained architecture
- Fix RL parameter check
- Split dimension check of actions in actor networks to new file
- Add some more testsEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/21Tensorflow group2019-09-25T16:17:50+02:00Julian Johannes Steinsberger-DührßenTensorflow groupMerge of the work done for the tensorflow backend during the autonomous driving lab SS 2019.Merge of the work done for the tensorflow backend during the autonomous driving lab SS 2019.Evgeny KusmenkoJulian Johannes Steinsberger-DührßenEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/20Integrate TD3 Algorithm and Gaussian Noise2019-08-05T13:29:42+02:00Nicola GattoIntegrate TD3 Algorithm and Gaussian NoiseChanges to Reinforcement Learning Parameters
- Add TD3 RL algorithm option
- Add TD3 related hyperparameters
- Add Gaussian strategy for exploration
- Make NNArchitecture a symbol for critic and trained architecture
- Add CoCos rela...Changes to Reinforcement Learning Parameters
- Add TD3 RL algorithm option
- Add TD3 related hyperparameters
- Add Gaussian strategy for exploration
- Make NNArchitecture a symbol for critic and trained architecture
- Add CoCos related to critic and trained architectureEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/19Fixed common-monticar version number2019-07-08T11:15:49+02:00Sebastian NickelsFixed common-monticar version numberAligned it with version number in other projects (CNNArchLang, EmbeddedMontiArcDL, ..) because it threw some errors.Aligned it with version number in other projects (CNNArchLang, EmbeddedMontiArcDL, ..) because it threw some errors.Evgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/18Feature/loss function2019-06-24T17:44:15+02:00Eyüp HarputluFeature/loss functionadded loss functionsadded loss functionsEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/17Add ddpg support2019-06-04T18:27:25+02:00Nicola GattoAdd ddpg support- Update the version to 0.3.2-SNAPSHOT
- Rename parameter action_selection to strategy
- Add new parameters for DDPG (see README)
- Change Common-monticar version to 0.0.17-SNAPSHOT, because version 0.0.17-20180824.094114-1 seems to b...- Update the version to 0.3.2-SNAPSHOT
- Rename parameter action_selection to strategy
- Add new parameters for DDPG (see README)
- Change Common-monticar version to 0.0.17-SNAPSHOT, because version 0.0.17-20180824.094114-1 seems to be non-existentEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/16Prepare ddpg algorithm2019-06-04T18:27:25+02:00Nicola GattoPrepare ddpg algorithm- Update the version to 0.3.1-SNAPSHOT
- Prepares configuration symbol and parameters for different rl algorithm:
- Add reinforcement learning algorithm parameter to switch between dqn and ddpg
- Add parameter for critic network
- ...- Update the version to 0.3.1-SNAPSHOT
- Prepares configuration symbol and parameters for different rl algorithm:
- Add reinforcement learning algorithm parameter to switch between dqn and ddpg
- Add parameter for critic network
- Annotate configuration symbol with information about the trained NN architectureEvgeny KusmenkoEvgeny Kusmenkohttps://git.rwth-aachen.de/monticore/EmbeddedMontiArc/languages/CNNTrainLang/-/merge_requests/15Add reinforcement learning2019-05-15T21:57:01+02:00Nicola GattoAdd reinforcement learning* Update the version to 0.3.0-SNAPSHOT
* Implement parameters for reinforcement learning (see README.md for detailed parameter information)
* Add more loss function parameters
* Add a convenient way to use multi parameter entries (see Mu...* Update the version to 0.3.0-SNAPSHOT
* Implement parameters for reinforcement learning (see README.md for detailed parameter information)
* Add more loss function parameters
* Add a convenient way to use multi parameter entries (see MultiParamValueSymbol)
* Add a RewardFunctionSymbol
* Add CoCo:
* From now on there are parameters exclusively for reinforcement learning and supervised learning. A check ensures that a parameter is used in the correct learning method context.
* If target network is used, a training interval is required
* Check that a reinforcement configuration always has information about the environment
* Given the ROS environment, a check ensures that a reward function is givenEvgeny KusmenkoEvgeny Kusmenko