From a0aa5d2ab4e41cc5cc843b02f3ecfbe65c0174b1 Mon Sep 17 00:00:00 2001 From: Grigorios Kalliatakis Date: Fri, 11 May 2018 09:25:27 +0100 Subject: [PATCH] Add known issues section for issue #5 --- README.md | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index 257ce62..bce2d16 100644 --- a/README.md +++ b/README.md @@ -1,10 +1,10 @@ -# Keras | VGG16 Places365 - VGG16 CNN models pre-trained on [Places365-Standard](http://places2.csail.mit.edu/download.html) for scene classification +# Keras | VGG16 Places365 - VGG16 CNN models pre-trained on Places365-Standard for scene classification ![Keras logo](https://i.imgur.com/c9r5WFp.png) [![license](https://img.shields.io/github/license/mashape/apistatus.svg?maxAge=2592000)](https://github.com/GKalliatakis/Keras-VGG16-places365/blob/master/LICENSE) -## You have just found the Keras models of the pre-trained VGG16 CNNs on Places365-Standard (~1.8 million images from 365 scene categories). +## You have just found the Keras models of the pre-trained VGG16 CNNs on [Places365-Standard](http://places2.csail.mit.edu/download.html) (~1.8 million images from 365 scene categories). ### Overview @@ -32,8 +32,6 @@ All architectures are compatible with both TensorFlow and Theano, and upon insta Pre-trained weights can be automatically loaded upon instantiation (`weights='places'` argument in model constructor for all image models). Weights are automatically downloaded. - - ## Examples ### Classify Places classes with VGG16-places365 @@ -122,6 +120,9 @@ More info on downloading, converting, and submitting other models can be found o ### Questions and Comments If you have any suggestions or bugs to report you can pull a request or start a discussion. + +### Known issues: +- Identical & low confidence predictions when classifying Places classes, as described in [issue #5](https://github.com/GKalliatakis/Keras-VGG16-places365/issues/5) _________________________________________________________________ [dill]: