luz
luz copied to clipboard
mlverse
Higher Level API for torch
After training a model using the luz_callback_model_checkpoint() function then loading the checkpoint using luz_load_checkpoint(), I am receiving the following error: Error in cpp_tensor_from_buffer(buffer, shape, list(dtype = dtype)) : Not compatible...
Hello, I just upgraded to macOS Sonoma 14.4.1, and it appears to have caused a breaking change with the 'luz' setup for multi-class models. Everything was running fine before, and...
I have installed TORCH correctly (it runs just fine) and I could even see a statement (during its reinstallations to solve the LUZ non-zero exit status) something like TORCH being...
- [x] AUC - [ ] Precision - [ ] Recall - [x] MSE - [x] RMSE - [x] MAE - [ ] Cohen's Kappa
Weights
We should provide a way for users to provide case weights as arguments for the loss function.
Question: How do I get the embeddings after fitting using triplet loss in this example: https://mlverse.github.io/luz/articles/examples/mnist-triplet.html ?
THis callback will trace the model for training and validation. Hopefully the optimizations that happen on tracing can generate some speedups for training more complicated models. TODO: - [ ]...
Related to https://github.com/mlverse/luz/issues/5#issuecomment-884426442 Hi @dfalbel, This is a first attempt at implementing the validation check callback. It may still need some work so I am submitting this as a draft...
- [x] Model checkpoint callback - [x] Early stopping - [x] Learning rate schedulers - [x] ndjson/csv logger - [ ] Gradient clipping