caffe学习笔记（4）：lr_policy之各模型形式的总结

最新推荐文章于 2020-02-13 15:21:45 发布

原创最新推荐文章于 2020-02-13 15:21:45 发布 · 6.4k 阅读

4 ·

CC 4.0 BY-SA版权

文章标签：

#lr-policy #cnn #caffe

deep learning 专栏收录该内容

16 篇文章

订阅专栏

本文总结了Caffe中针对AlexNet、GoogleNet、CaffeNet和CIFAR10模型的lr_policy，分别介绍了各模型的特点和实现细节。AlexNet模型遵循原始论文描述；GoogleNet模型得益于Christian Szegedy的帮助得以复现；CaffeNet在AlexNet基础上略有调整；CIFAR10模型则根据Alex Krizhevsky的cuda-convnet教程在Caffe中实现。

部署运行你感兴趣的模型镜像

alexnet模型

net: "models/bvlc_alexnet/train_val.prototxt"
test_iter: 1000
test_interval: 1000
base_lr: 0.01
lr_policy: "step"
gamma: 0.1
stepsize: 100000
display: 20
max_iter: 450000
momentum: 0.9
weight_decay: 0.0005
snapshot: 10000
snapshot_prefix: "models/bvlc_alexnet/caffe_alexnet_train"
solver_mode: GPU

This model is a replication of the model described in the AlexNet publication.

googlenet模型

net: "models/bvlc_googlenet/train_val.prototxt"
test_iter: 1000
test_interval: 4000
test_initialization: false
display: 40
average_loss: 40
base_lr: 0.01
lr_policy: "step"
stepsize: 320000
gamma: 0.96
max_iter: 10000000
momentum: 0.9
weight_decay: 0.0002
snapshot: 40000
snapshot_prefix: "models/bvlc_googlenet/bvlc_googlenet"
solver_mode: GPU

This model is a replication of the model described in the GoogleNet publication. We would like to thank Christian Szegedy for all his help in the replication of GoogleNet model.

caffenet模型

net: "models/bvlc_reference_caffenet/train_val.prototxt"
test_iter: 1000
test_interval: 1000
base_lr: 0.01
lr_policy: "step"
gamma: 0.1
stepsize: 100000
display: 20
max_iter: 450000
momentum: 0.9
weight_decay: 0.0005
snapshot: 10000
snapshot_prefix: "models/bvlc_reference_caffenet/caffenet_train"
# solver_mode: GPU

This model is the result of following the Caffe ImageNet model training instructions.
It is a replication of the model described in the AlexNet publication with some differences:

not training with the relighting data-augmentation;
the order of pooling and normalization layers is switched (in CaffeNet, pooling is done before normalization).

cifar10模型

# The train/test net protocol buffer definition
net: "examples/cifar10/cifar10_quick_train_test.prototxt"
# test_iter specifies how many forward passes the test should carry out.
# In the case of MNIST, we have test batch size 100 and 100 test iterations,
# covering the full 10,000 testing images.
test_iter: 100
# Carry out testing every 500 training iterations.
test_interval: 500
# The base learning rate, momentum and the weight decay of the network.
base_lr: 0.001
momentum: 0.9
weight_decay: 0.004
# The learning rate policy
lr_policy: "fixed"
# Display every 100 iterations
display: 100
# The maximum number of iterations
max_iter: 4000
# snapshot intermediate results
snapshot: 4000
snapshot_format: HDF5
snapshot_prefix: "examples/cifar10/cifar10_quick"
# solver mode: CPU or GPU
solver_mode: GPU

Alex’s CIFAR-10 tutorial, Caffe style

Alex Krizhevsky’s cuda-convnet details the model definitions, parameters, and training procedure for good performance on CIFAR-10. This example reproduces his results in Caffe.

We will assume that you have Caffe successfully compiled. If not, please refer to the Installation page. In this tutorial, we will assume that your caffe installation is located at CAFFE_ROOT.

We thank @chyojn for the pull request that defined the model schemas and solver configurations.

This example is a work-in-progress. It would be nice to further explain details of the network and training choices and benchmark the full training.

未完待续。。

您可能感兴趣的与本文相关的镜像