search

TuSimple / TuSimple-DUC

Understanding Convolution for Semantic Segmentation
https://arxiv.org/abs/1702.08502
Apache License 2.0
610 stars 118 forks source link

Where is the ImageNet pretrained classification models? #1

Closed yzou2 closed 7 years ago

yzou2 commented 7 years ago

In the link to the models (https://drive.google.com/drive/folders/0B72xLTlRb0SoREhISlhibFZTRmM), I can only find the segmentation pretrained models. Would you please tell me where can I find the classification model pretrained on ImageNet?

wpqmanu commented 7 years ago

You could use any publicly available pretrained ResNet model using MXNet, such as the ones here. You can refer to our symbol to match the name of the layers. Or to get a quicker start, you could use the pretrained DeepLab V2 caffe model and convert it to the MXNet format.

shipengai commented 6 years ago

Hi, I didn't install caffe . For quickly to train, can anyone share the pretrained Resnet model that converted from deeplab v2 caffe model?@yzou2@wpqmanu Thanks. One more question.How to match the name of layers of your symbol? Thanks.

shipengai commented 6 years ago

Hi , I found init.params. Is it the pretrained Resnet model that converted from deeplab v2 caffe model? @wpqmanu

wpqmanu commented 6 years ago

Yes

On Feb 26, 2018 19:18, "shipeng" notifications@github.com wrote:

Hi , I found init.params. Is it the pretrained Resnet model that converted from deeplab v2 caffe model? @wpqmanu https://github.com/wpqmanu

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/TuSimple/TuSimple-DUC/issues/1#issuecomment-368733735, or mute the thread https://github.com/notifications/unsubscribe-auth/ACXJi6NWAjZ9SPAsElkKU4jJfds0hT7kks5tY3QYgaJpZM4QXYaW .