IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an email to s dot adaszewski at gmail dot com. User accounts are meant only to report issues and/or generate pull requests. This is a purpose-specific Git hosting for ADARED projects. Thank you for your understanding!
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Stanislaw Adaszewski 2eff854467 Start working on FastLoop. пре 3 година
..
unused Remove loss as no longer necessary. пре 3 година
__init__.py Work on icosagon.trainprep. пре 3 година
batch.py Introduce FlatPredictions, flatten_predictions(), BatchIndices, gather_batch_indices(). пре 3 година
bulkdec.py Fix a typo. пре 3 година
compile.py Performance tests. пре 3 година
convlayer.py Add BatchedData support to DecodeLayer. пре 3 година
convolve.py Remove an extra newline. пре 3 година
data.py Clean up _equal(). пре 3 година
databatch.py Add BatchedData support to DecodeLayer. пре 3 година
declayer.py Add BatchedData support to DecodeLayer. пре 3 година
decode.py Use torch.nn.Parameter(List) in decode. пре 3 година
dropout.py Remove unnecessary .detach() call in dropout_dense(). пре 3 година
fastconv.py Reimplement FastConvLayer. пре 3 година
fastloop.py Start working on FastLoop. пре 3 година
fastmodel.py Add test_fast_model_01(). пре 3 година
input.py Fix for non-leaf .grad warning. пре 3 година
model.py Add TrainLoop. пре 3 година
normalize.py Fixes for bool sparse tensors. пре 3 година
sampling.py Fixes for the GPU. пре 3 година
trainloop.py Add some debug output for profiling, the bottleneck is in DecodeLayer but also comes generally from computing always all nodes. пре 3 година
trainprep.py Fixes for the GPU. пре 3 година
weights.py Bring more elements to icosagon. пре 3 година