-
Dear Team,
I was reading through the codes of KernelGAN and I am interested in adopting the code for :
1. Monocolor (1 Channel) Image.
2. Hyperspectral Images (More than 3 color channels).
Could p…
-
I understand that FSRS v5 won't be released for a while, but I still decided to open this issue to collect good ideas.
There is already [an issue for D](https://github.com/open-spaced-repetition/fsrs…
-
Hi @kalebentley, sorry this has taken so long. The data wrangling turned out to be nontrivial; I kept encountering unanticipated wrinkles and did a lot of iterating between data cleaning and model fit…
-
Hi,
Thanks for sharing your work !
I would like what is your configuration for the training of yolov3.cfg to get 55% MAP ?
We tried 100 epochs but we got a MAP (35%) who don't really change much …
-
Lucene/Solr can host huge sets of documents containing lots of information in fields so that these can be used as training examples (w/ features) in order to very quickly create classifiers algorithms…
-
Song Y, Ermon S. [Generative modeling by estimating gradients of the data distribution](https://proceedings.neurips.cc/paper/2019/file/3001ef257407d5a371a96dcd947c7d93-Paper.pdf).
Yang Song. [Generat…
-
Following from a discussion on Slack:
> One thing which I think we need to solve is that there is a key architectural difference between Soss/Gen and Turing.
In the former case, you can construct …
-
Hello, all!
I was speaking with @yebai about how Turing could support the whole Bayesian Workflow following Bob Carpenter presentation at ProbProg 2021 in October, 2021. And I gave him an idea abou…
-
In `v3` we (ab)used the Theano `.tag.test_value` to evaluate the model (`Model.check_test_point`) at a set point as well as provide a starting point for the sampler, this point was stored in the `Mode…
-
I am trying to use a multitask gaussian process for Bayesian optimization with a dataset that has about 10,000 records. In order to speedup training, I am trying to use mini-batch SGD to find the hype…