跳转到主要内容

一个基于PyTorch的软件平台,用于在普渡大学教授深度学习课程

项目描述

有关此模块的所有信息,包括代码的最新更改信息,请参阅

https://engineering.purdue.edu/kak/distDLS/DLStudio-2.5.1.html

的相关模块API页面。

convo_layers_config = "1x[128,3,3,1]-MaxPool(2) 1x[16,5,5,1]-MaxPool(2)"
fc_layers_config = [-1,1024,10]

dls = DLStudio(
                  dataroot = "/home/kak/ImageDatasets/CIFAR-10/",
                  image_size = [32,32],
                  convo_layers_config = convo_layers_config,
                  fc_layers_config = fc_layers_config,
                  path_saved_model = "./saved_model",
                  momentum = 0.9,
                  learning_rate = 1e-3,
                  epochs = 2,
                  batch_size = 4,
                  classes = ('plane','car','bird','cat','deer','dog','frog','horse','ship','truck'),
                  use_gpu = True,
                  debug_train = 0,
                  debug_test = 1
              )

configs_for_all_convo_layers = dls.parse_config_string_for_convo_layers()
convo_layers = dls.build_convo_layers2( configs_for_all_convo_layers )
fc_layers = dls.build_fc_layers()
model = dls.Net(convo_layers, fc_layers)
dls.show_network_summary(model)
dls.load_cifar_10_dataset()
dls.run_code_for_training(model)
dls.run_code_for_testing(model)

项目详情


下载文件

下载适合您平台的文件。如果您不确定选择哪一个,请了解更多关于 安装软件包 的信息。

源分发

DLStudio-2.5.1.tar.gz (366.5 kB 查看哈希值)

上传时间

由以下支持