贝叶斯神经网络:Bayprop By Backprop,MC Dropout,SGLD,本地重新参数化技巧,KF-Laplace,SG-HMC的Pytorch实现-源码

上传者: 42117150 | 上传时间: 2021-03-28 22:19:28 | 文件大小: 13.93MB | 文件类型: ZIP
贝叶斯神经网络 以下近似推理方法的Pytorch实现: 我们还提供以下代码: 先决条件 火炬 脾气暴躁的 Matplotlib 该项目是用python 2.7和Pytorch 1.0.1编写的。 如果CUDA可用,它将自动使用。 这些模型也不会太大,因此也可以在CPU上运行。 用法 结构体 回归实验 我们对用 生成的玩具数据集和真实数据(六个)进行了均方差和异方差回归实验。 Notebooks / classification /(ModelName)_(ExperimentType).ipynb :包含在(ExperimentType)上使用(ModelName)进行的实验,即同调/异

文件下载

资源详情

[{"title":"( 74 个子文件 13.93MB ) 贝叶斯神经网络:Bayprop By Backprop,MC Dropout,SGLD,本地重新参数化技巧,KF-Laplace,SG-HMC的Pytorch实现-源码","children":[{"title":"Bayesian-Neural-Networks-master","children":[{"title":"notebooks","children":[{"title":"regression","children":[{"title":"langevin_homo.ipynb <span style='color:#111;'> 99.81KB </span>","children":null,"spread":false},{"title":"langevin_hetero.ipynb <span style='color:#111;'> 185.55KB </span>","children":null,"spread":false},{"title":"bbp_homo.ipynb <span style='color:#111;'> 98.54KB </span>","children":null,"spread":false},{"title":"map_hetero.ipynb <span style='color:#111;'> 128.13KB </span>","children":null,"spread":false},{"title":"map_homo.ipynb <span style='color:#111;'> 112.60KB </span>","children":null,"spread":false},{"title":"bbp_hetero.ipynb <span style='color:#111;'> 550.13KB </span>","children":null,"spread":false},{"title":"mc_dropout_homo.ipynb <span style='color:#111;'> 70.24KB </span>","children":null,"spread":false},{"title":"gp_homo_hetero.ipynb <span style='color:#111;'> 177.49KB </span>","children":null,"spread":false},{"title":"mc_dropout_hetero.ipynb <span style='color:#111;'> 134.94KB </span>","children":null,"spread":false}],"spread":true},{"title":"classification","children":[{"title":"KFAC_Laplace_MNIST.ipynb <span style='color:#111;'> 535.48KB </span>","children":null,"spread":false},{"title":"weight_uncertainty_local_reparam_MNIST.ipynb <span style='color:#111;'> 708.25KB </span>","children":null,"spread":false},{"title":"MAP_NN_MNIST.ipynb <span style='color:#111;'> 531.69KB </span>","children":null,"spread":false},{"title":"pSGLD_MNIST.ipynb <span style='color:#111;'> 619.02KB </span>","children":null,"spread":false},{"title":"weight_uncertainty_MC_MNIST_importance_weigh.ipynb <span style='color:#111;'> 965.71KB </span>","children":null,"spread":false},{"title":"Bootstrap_ensemble_NN_MNIST.ipynb <span style='color:#111;'> 6.02MB </span>","children":null,"spread":false},{"title":"PLOTS.ipynb <span style='color:#111;'> 1.16MB </span>","children":null,"spread":false},{"title":"BayesByBackprop_MNIST_GMM.ipynb <span style='color:#111;'> 633.34KB </span>","children":null,"spread":false},{"title":"MC_dropout_MNIST.ipynb <span style='color:#111;'> 557.50KB </span>","children":null,"spread":false},{"title":"SGLD_MNIST.ipynb <span style='color:#111;'> 683.95KB </span>","children":null,"spread":false},{"title":"BayesByBackprop_MNIST_gaussian.ipynb <span style='color:#111;'> 954.88KB </span>","children":null,"spread":false},{"title":"weight_uncertainty_MC_MNIST_laplace.ipynb <span style='color:#111;'> 913.27KB </span>","children":null,"spread":false}],"spread":false}],"spread":true},{"title":"train_BootrapEnsemble_MNIST.py <span style='color:#111;'> 8.68KB </span>","children":null,"spread":false},{"title":"src","children":[{"title":"Stochastic_Gradient_HMC_SA","children":[{"title":"model.py <span style='color:#111;'> 6.09KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"optimizers.py <span style='color:#111;'> 4.39KB </span>","children":null,"spread":false}],"spread":true},{"title":"MC_dropout","children":[{"title":"model.py <span style='color:#111;'> 5.72KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false}],"spread":true},{"title":"Bayes_By_Backprop","children":[{"title":"model.py <span style='color:#111;'> 15.15KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false}],"spread":true},{"title":"priors.py <span style='color:#111;'> 1.90KB </span>","children":null,"spread":false},{"title":"utils.py <span style='color:#111;'> 4.01KB </span>","children":null,"spread":false},{"title":"Bootstrap_Ensemble","children":[{"title":"model.py <span style='color:#111;'> 5.66KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false}],"spread":true},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"Bayes_By_Backprop_Local_Reparametrization","children":[{"title":"model.py <span style='color:#111;'> 14.68KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false}],"spread":true},{"title":"base_net.py <span style='color:#111;'> 1.32KB </span>","children":null,"spread":false},{"title":"Stochastic_Gradient_Langevin_Dynamics","children":[{"title":"model.py <span style='color:#111;'> 6.42KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"optimizers.py <span style='color:#111;'> 4.01KB </span>","children":null,"spread":false}],"spread":true},{"title":"KF_Laplace","children":[{"title":"hessian_operations.py <span style='color:#111;'> 2.16KB </span>","children":null,"spread":false},{"title":"model.py <span style='color:#111;'> 10.66KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false}],"spread":false}],"spread":false},{"title":"images","children":[{"title":"MAP.png <span style='color:#111;'> 270.02KB </span>","children":null,"spread":false},{"title":"uci_regression_log_liks.png <span style='color:#111;'> 229.06KB </span>","children":null,"spread":false},{"title":"all_rotations.png <span style='color:#111;'> 518.70KB </span>","children":null,"spread":false},{"title":"adv_entropy.png <span style='color:#111;'> 518.17KB </span>","children":null,"spread":false},{"title":"uci_regression_rmses.png <span style='color:#111;'> 217.22KB </span>","children":null,"spread":false},{"title":"map_homo.png <span style='color:#111;'> 141.23KB </span>","children":null,"spread":false},{"title":"gp_hetero.png <span style='color:#111;'> 112.02KB </span>","children":null,"spread":false},{"title":"uci_regression_rmses.pdf <span style='color:#111;'> 26.71KB </span>","children":null,"spread":false},{"title":"map_hetero.png <span style='color:#111;'> 119.50KB </span>","children":null,"spread":false},{"title":"mc_dropout_homo.png <span style='color:#111;'> 127.14KB </span>","children":null,"spread":false},{"title":"BBP_train.png <span style='color:#111;'> 646.07KB </span>","children":null,"spread":false},{"title":"Wdistribution.png <span style='color:#111;'> 143.96KB </span>","children":null,"spread":false},{"title":"mc_dropout_hetero.png <span style='color:#111;'> 118.76KB </span>","children":null,"spread":false},{"title":"sgld_homo.png <span style='color:#111;'> 124.08KB </span>","children":null,"spread":false},{"title":"heteroscedastic_regression.png <span style='color:#111;'> 347.97KB </span>","children":null,"spread":false},{"title":"uci_regression_log_liks.pdf <span style='color:#111;'> 27.88KB </span>","children":null,"spread":false},{"title":"KMNIST_entropies.png <span style='color:#111;'> 281.11KB </span>","children":null,"spread":false},{"title":"bbp_homo.png <span style='color:#111;'> 114.34KB </span>","children":null,"spread":false},{"title":"sgld_hetero.png <span style='color:#111;'> 96.06KB </span>","children":null,"spread":false},{"title":"gp_homo.png <span style='color:#111;'> 103.74KB </span>","children":null,"spread":false},{"title":"bbp_hetero.png <span style='color:#111;'> 102.91KB </span>","children":null,"spread":false},{"title":"homoscedastic_regression.png <span style='color:#111;'> 391.71KB </span>","children":null,"spread":false},{"title":"adv_robustness.png <span style='color:#111;'> 336.67KB </span>","children":null,"spread":false}],"spread":false},{"title":"train_MCDropout_MNIST.py <span style='color:#111;'> 7.57KB </span>","children":null,"spread":false},{"title":"train_SGHMC_MNIST.py <span style='color:#111;'> 7.75KB </span>","children":null,"spread":false},{"title":"train_BayesByBackprop_MNIST.py <span style='color:#111;'> 10.30KB </span>","children":null,"spread":false},{"title":"LICENSE <span style='color:#111;'> 1.05KB </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 14.30KB </span>","children":null,"spread":false},{"title":"train_SGLD_MNIST.py <span style='color:#111;'> 8.42KB </span>","children":null,"spread":false},{"title":".gitignore <span style='color:#111;'> 4.03KB </span>","children":null,"spread":false},{"title":"train_KFLaplace_MNIST.py <span style='color:#111;'> 9.01KB </span>","children":null,"spread":false}],"spread":false}],"spread":true}]

评论信息

免责申明

【只为小站】的资源来自网友分享,仅供学习研究,请务必在下载后24小时内给予删除,不得用于其他任何用途,否则后果自负。基于互联网的特殊性,【只为小站】 无法对用户传输的作品、信息、内容的权属或合法性、合规性、真实性、科学性、完整权、有效性等进行实质审查;无论 【只为小站】 经营者是否已进行审查,用户均应自行承担因其传输的作品、信息、内容而可能或已经产生的侵权或权属纠纷等法律责任。
本站所有资源不代表本站的观点或立场,基于网友分享,根据中国法律《信息网络传播权保护条例》第二十二条之规定,若资源存在侵权或相关问题请联系本站客服人员,zhiweidada#qq.com,请把#换成@,本站将给予最大的支持与配合,做到及时反馈和处理。关于更多版权及免责申明参见 版权及免责申明