juice
juice copied to clipboard
fff-rs
The Hacker's Machine Learning Engine
…reshape inputs. Update examples to use isize and variable batch_size. ## What does this PR accomplish? * 🦚 Feature ## Changes proposed by this PR: ## Notes to reviewer: ##...
Use case: I’ve written a [module](https://github.com/TheodoreEhrenborg/phutball/) in Python that plays a board game. Part of the module uses a convolutional neural network, which I made with TensorFlow Keras. I’d like...
## What does this PR accomplish? * 🦚 Feature Closes #145. ## Changes proposed by this PR: - [ ] Adding the necessary wrapper functions of the ffi::* cuDNN BatchNormaliztion...
Hi @drahnr, I'd like to take a crack at implementing the BatchNormalization2d layer in juice. Referring back to #10 you outlined that adding new cuDNN layers to juice is a...
Move to `thiserror` to simplify a few things around errors in the long run. Also deals with the deprecation of `description` in favour of `fmt::Display` and such.
**Describe the bug** Currently the rnn mackey glass example under `./juice-examples` does not work, if `train` and `test` are run with different `--batch-size` arguments. **To Reproduce** Steps to reproduce the...
### Summary RNN Forward fails with CUDNN RNN when using Bidirectional as a direction mode as compared to Unidirectional. ### Code Snippet ``` net_cfg.add_layer(LayerConfig::new( // Layer name is only used...
currently everything is recompiled + fetched from remote which is not necessary * [ ] figure out how to use `sccache` from `concourse` without exposing the host network, probably via...
Bigger work items: * [x] Recursive Neural Networks / Long Term Short Term Memory #87 * [ ] OpenCL Backend [juice/coaster-nn] * [ ] Autodiff [juice/coaster-nn] * [ ] Accuracy...
Network being used is LSTM -> Linear -> Sigmoid network using Mackey-Glass. ``` net_cfg.add_layer(LayerConfig::new( // Layer name is only used internally - can be changed to anything "LSTMInitial", RnnConfig {...