Skip to content

Conversation

ailzhang
Copy link
Contributor

This PR add a few resnext, segmentation models, shufflenext, googlenet, mobilenet in hubconf.
Here is a few quick experiments:

In [3]: torch.hub.list('ailzhang/vision:add_segmentation_hubconf', force_reload=True)
Downloading: "https://github.com/ailzhang/vision/archive/add_segmentation_hubconf.zip" to /private/home/ailzhang/.cache/torch/hub/add_segmentation_hubconf.zip
Out[3]:
['alexnet',
 'deeplabv3_resnet101',
 'densenet121',
 'densenet161',
 'densenet169',
 'densenet201',
 'fcn_resnet101',
 'googlenet',
 'inception_v3',
 'mobilenet_v2',
 'resnet101',
 'resnet152',
 'resnet18',
 'resnet34',
 'resnet50',
 'resnext101_32x8d',
 'resnext50_32x4d',
 'shufflenet_v2_x0_5',
 'shufflenet_v2_x1_0',
 'squeezenet1_0',
 'squeezenet1_1',
 'vgg11',
 'vgg11_bn',
 'vgg13',
 'vgg13_bn',
 'vgg16',
 'vgg16_bn',
 'vgg19',
 'vgg19_bn']


In [4]: model = torch.hub.load('ailzhang/vision:add_segmentation_hubconf', 'mobilenet_v2', pretrained=True)
Using cache found in /private/home/ailzhang/.cache/torch/hub/ailzhang_vision_add_segmentation_hubconf
Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /private/home/ailzhang/.cache/torch/checkpoints/mobilenet_v2-b0353104.pth
100%|██████████████████████████████████████████████████████████████████████████████████████████████| 13.6M/13.6M [00:00<00:00, 19.3MB/s]

In [5]: model = torch.hub.load('ailzhang/vision:add_segmentation_hubconf', 'googlenet', pretrained=True)
Using cache found in /private/home/ailzhang/.cache/torch/hub/ailzhang_vision_add_segmentation_hubconf
Downloading: "https://download.pytorch.org/models/googlenet-1378be20.pth" to /private/home/ailzhang/.cache/torch/checkpoints/googlenet-1378be20.pth
100%|██████████████████████████████████████████████████████████████████████████████████████████████| 49.7M/49.7M [00:01<00:00, 35.4MB/s]

In [6]: model = torch.hub.load('ailzhang/vision:add_segmentation_hubconf', 'resnext101_32x8d', pretrained=True)
Using cache found in /private/home/ailzhang/.cache/torch/hub/ailzhang_vision_add_segmentation_hubconf
Downloading: "https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth" to /private/home/ailzhang/.cache/torch/checkpoints/resnext101_32x8d-8ba56ff5.pth
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 340M/340M [00:08<00:00, 40.7MB/s]

In [7]: model = torch.hub.load('ailzhang/vision:add_segmentation_hubconf', 'resnext50_32x4d', pretrained=True)
Using cache found in /private/home/ailzhang/.cache/torch/hub/ailzhang_vision_add_segmentation_hubconf
Downloading: "https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth" to /private/home/ailzhang/.cache/torch/checkpoints/resnext50_32x4d-7cdf4587.pth
100%|██████████████████████████████████████████████████████████████████████████████████████████████| 95.8M/95.8M [00:03<00:00, 29.8MB/s]

In [8]: model = torch.hub.load('ailzhang/vision:add_segmentation_hubconf', 'shufflenet_v2_x0_5', pretrained=True)
Using cache found in /private/home/ailzhang/.cache/torch/hub/ailzhang_vision_add_segmentation_hubconf
Downloading: "https://download.pytorch.org/models/shufflenetv2_x0.5-f707e7126e.pth" to /private/home/ailzhang/.cache/torch/checkpoints/shufflenetv2_x0.5-f707e7126e.pth
100%|██████████████████████████████████████████████████████████████████████████████████████████████| 5.28M/5.28M [00:00<00:00, 9.80MB/s]

In [9]: model = torch.hub.load('ailzhang/vision:add_segmentation_hubconf', 'shufflenet_v2_x1_0', pretrained=True)
Using cache found in /private/home/ailzhang/.cache/torch/hub/ailzhang_vision_add_segmentation_hubconf
Downloading: "https://download.pytorch.org/models/shufflenetv2_x1-5666bf0f80.pth" to /private/home/ailzhang/.cache/torch/checkpoints/shufflenetv2_x1-5666bf0f80.pth
100%|██████████████████████████████████████████████████████████████████████████████████████████████| 8.79M/8.79M [00:00<00:00, 15.7MB/s]

In [2]: model = torch.hub.load('ailzhang/vision:add_segmentation_hubconf', 'deeplabv3_resnet101', pretrained=True)
Using cache found in /private/home/ailzhang/.cache/torch/hub/ailzhang_vision_add_segmentation_hubconf

In [3]: model = torch.hub.load('ailzhang/vision:add_segmentation_hubconf', 'fcn_resnet101', pretrained=True)
Using cache found in /private/home/ailzhang/.cache/torch/hub/ailzhang_vision_add_segmentation_hubconf

Copy link
Member

@fmassa fmassa left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, thanks!

Can you just fix the lint? Then this is good to merge

@codecov-io
Copy link

codecov-io commented May 31, 2019

Codecov Report

Merging #974 into master will not change coverage.
The diff coverage is n/a.

Impacted file tree graph

@@           Coverage Diff           @@
##           master     #974   +/-   ##
=======================================
  Coverage   61.16%   61.16%           
=======================================
  Files          65       65           
  Lines        5091     5091           
  Branches      764      764           
=======================================
  Hits         3114     3114           
  Misses       1767     1767           
  Partials      210      210

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 220b69b...ba64370. Read the comment docs.

@soumith soumith merged commit 47db963 into pytorch:master May 31, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants