From 42b09524b1bce788e7bcc190a55a6f020895fba8 Mon Sep 17 00:00:00 2001 From: Peter Kontschieder Date: Thu, 14 Feb 2019 09:25:00 +0100 Subject: [PATCH] Adding pretrained ResNet34/50 models --- README.md | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 4b50738..8124fe7 100644 --- a/README.md +++ b/README.md @@ -21,8 +21,8 @@ We have now also released the inference code for semantic segmentation, together - **Added compatibility with fp16 (currently allows fp16 input but requires the module to stay in fp32 mode)** - **Requires now PyTorch 1.0** -**Update 12 Feb. 2019:** -- **Added ResNet models** +**Update 14 Feb. 2019:** +- **Added ResNet34 and ResNet50 models** We have modified the imagenet training code and BN synchronization in order to work with multiple processes. We have also added compatibility of our Inplace ABN module with fp16. @@ -80,6 +80,10 @@ Here you can find the results from our arXiv paper (top-1 / top-5 scores) with c | [WideResNet38, InPlace-ABN][4] | 256 | 79.72 / 94.78 | 81.03 / 95.43 | 80.69 / 95.27 | [`1c085ab70b789cc1d6c1594f7a761007`][9] | | [ResNeXt101, InPlace-ABN sync][5] | 256 | 77.70 / 93.78 | 79.18 / 94.60 | 78.98 / 94.56 | [`0a85a21847b15e5a242e17bf3b753849`][10] | | [DenseNet264, InPlace-ABN][11] | 256 | 78.57 / 94.17 | 79.72 / 94.93 | 79.49 / 94.89 | [`0b413d67b725619441d0646d663865bf`][12] | +| [ResNet50, InPlace-ABN sync][13] | 512 | 75.53 / 92.59 | 77.04 / 93.57 | 76.60 / 93.49 | [`2522ca639f7fdfd7c0089ba1f5f6c2e8`][14] | +| [ResNet34, InPlace-ABN sync][15] | 512 | 73.27 / 91.34 | 75.19 / 92.66 | 74.87 / 92.42 | [`61515c1484911c3cc753d405131e1dda`][16] | + + [1]: experiments/resnext101_stdbn_lr_256.json [2]: experiments/resnext101_ipabn_lr_512.json @@ -93,6 +97,10 @@ Here you can find the results from our arXiv paper (top-1 / top-5 scores) with c [10]: https://drive.google.com/file/d/1v2gmUPBMDKf0wZm9r1JwCQLGAig0DdXJ/view [11]: experiments/densenet264_ipabn_lr_256.json [12]: https://drive.google.com/file/d/1J2wp59bzzEd6zttM6oMa1KgbmCL1MS0k/view +[13]: experiments/resnet50_ipabn-sync_lr_512.json +[14]: https://drive.google.com/file/d/1N7kjWrnUbD_aBOUNi9ZLGnI3E_1ATH8U/view +[15]: experiments/resnet34_ipabn-sync_lr_512.json +[16]: https://drive.google.com/file/d/1V5dCIZeRCfnZi9krNaQNhXNDHyXz9JR8/view ### Data preparation Our script uses [torchvision.datasets.ImageFolder](http://pytorch.org/docs/master/torchvision/datasets.html#torchvision.datasets.ImageFolder)