DeastinY
Hi all, the tutorial contains the following section on this: > Checking how the model performs across different slices of the data allows us to detect weaknesses. Plotted below are...
If the prediction did not work correctly it would be nice if it was possible to tell so. The algorithm could learn from those false positives
Has this been discussed somewhere? This would allow to add many roasters in a non-invasive manner that were not originally designed to work with artisan.
With the current implementation (afaik) only a single step will be predicted. A modification for the Decoder part would be great
So based on #103 I downloaded release code from 2.6, but I end up with ``` -- The C compiler identification is GNU 7.4.0 -- The CXX compiler identification is...
does it work with and handle wine?
ASR
I've been using the ESPHome variant for a few days and today set up an automation to press M1 / M2 every 30 minutes. After some time it failed and...
Normally I'd fix them by adding ``` \clubpenalty10000 \widowpenalty10000 \displaywidowpenalty=10000 ``` but this screws with the references. Any other options ? Any idea why this messes with citations ? **Edit:**...
As DA-RNN is basically a encoder-decoder network, how would you do multi step prediction in the decoder phase?