decagon-pytorch

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!

Stanislaw Adaszewski 2eff854467 Start working on FastLoop.		3年前
..
unused	Remove loss as no longer necessary.	3年前
__init__.py	Work on icosagon.trainprep.	3年前
batch.py	Introduce FlatPredictions, flatten_predictions(), BatchIndices, gather_batch_indices().	3年前
bulkdec.py	Fix a typo.	3年前
compile.py	Performance tests.	3年前
convlayer.py	Add BatchedData support to DecodeLayer.	3年前
convolve.py	Remove an extra newline.	3年前
data.py	Clean up _equal().	3年前
databatch.py	Add BatchedData support to DecodeLayer.	3年前
declayer.py	Add BatchedData support to DecodeLayer.	3年前
decode.py	Use torch.nn.Parameter(List) in decode.	3年前
dropout.py	Remove unnecessary .detach() call in dropout_dense().	3年前
fastconv.py	Reimplement FastConvLayer.	3年前
fastloop.py	Start working on FastLoop.	3年前
fastmodel.py	Add test_fast_model_01().	3年前
input.py	Fix for non-leaf .grad warning.	3年前
model.py	Add TrainLoop.	3年前
normalize.py	Fixes for bool sparse tensors.	3年前
sampling.py	Fixes for the GPU.	3年前
trainloop.py	Add some debug output for profiling, the bottleneck is in DecodeLayer but also comes generally from computing always all nodes.	3年前
trainprep.py	Fixes for the GPU.	3年前
weights.py	Bring more elements to icosagon.	3年前