Cyrus Asfa comments

Results 14 comments of


                                            Cyrus Asfa

init() should return None, not 'NoneType' when instantiating RenderEngine()

I fixed this by changing the link flags in the build settings. Go to build settings in Xcode and change `-lpython2.7m` to `-undefined dynamic_lookup` in 'Other Linker Flags' Refer to...

Width/height setting of 0px causes errors on some interfaces

When creating a nexus object in a React `componentDidMount()` I get the following error > NexusUI.js:970 > Uncaught TypeError: Cannot set property 'width' of undefined at e.value (NexusUI.js:970) at e...

pairwise_dist drops to 0，loss is near the margin and can't go down

> Maybe pre-training with a softmax loss could help. Hi Olivier, I considered this approach to try learning some supervised representation from the data, then refining it with triplet learning....

pairwise_dist drops to 0，loss is near the margin and can't go down

> The pretraining approach is just to get a good embedding with a softmax loss, since this loss is very stable and you should be able to converge. > >...

pairwise_dist drops to 0，loss is near the margin and can't go down

Ok thanks. I more wanted to ask if the embedding's activation should be linear instead of ReLu, which I have seen mentioned before

Adding data pipeline for balanced batches

> If you are working with images stored in jpg files for instance, you can apply `tf.data.experimental.choose_from_datasets` only on the filenames and labels (which should be very fast), and then...

ENH Transpose backend primitive

> @MuawizChaudhary LMK if/when you need a new review now

Open source URMP dataset pipeline

looks good to me. only thing I notice is that `prepare_urmp_dataset_lib.parse_example` expects TFRecords. Will these `TFRecords` or the code to produce them also be available?

sphinx 1d/plot_filters figures don't match code

strange, i though #899 fixed this?

Bug in JTFS S1 shape and `subsample_fourier` during frequency scattering (numpy and tensorflow)

> these should all be aligned and the backend should regularize everything to numpy convention (which for stack means axis=0 by default) > […](#) Right. This turns out to not...