Daniel Stoller

Results 9 issues of Daniel Stoller

I have a blog post with some articles I reference, (see this URL https://dans.world/Spotify-internship/ ) Within the text, links get generated for the references to the citations, which get listed...

Webdataset does not list `torch` as dependency, so `torch` is not present when installing webdataset==0.2.5 with a non-Torch project. However, it then fails when doing ``` from webdataset import TarWriter...

add testcase

Hey! Thanks for your great work on this library. I encountered very weird training behaviour with wds. I generated my tar files all in one folder as follows: ``` training_faa0dfb1-9775-4279-8146-f251997958c4.tar...

documentation
errormessage

Hello again, thanks for providing the Youtube URLs. I just have a minor suggestion. You also annotated in ``youtube_urls.csv`` if the Youtube version was slightly slower or faster than the...

enhancement

In the paper you can see the diagram of Fixup which applies the residual branch, and adds the output to the original input. Finally it applies a ReLU to this...

Taken from issue #1 - @kalai2033 asks: Can I use a rectangular image without resizing? My i/p image size is 600*400.

question

quniform is still giving me a float value instead of an int, just like described in issue #253. It was supposedly fixed, but for me the problem persists. See the...

Stale

Percentage of correct segments is currently computed by also checking for overlap of predicted and actual silence durations between words. However for Mauch and Jamendo datasets there are no annotated...

Adjusts README to the fact that the paper got accepted to ICML and changed title slightly