lorenzoh

Results 41 issues of lorenzoh

Sometimes encodings need to be able to take into account batch information, as in a sequence learning task where samples in a batch should be padded to the length of...

api-proposal

CI fails on Windows at the moment due to file permission errors that occur when downloading datadeps. See for example this run: https://github.com/FluxML/FastAI.jl/runs/6899007003?check_suite_focus=true

In addition to existing feature registries, FastAI.jl will be getting a model registry. The model registry should allow - domain and third-party packages to add additional models when loaded; -...

enhancement
api-proposal
plans

This adds interface tests to various interfaces in FastAI.jl. This allows - giving user-friendly error messages by performing checks in high-level functions - giving help when implementing an interface, and...

Now that Optimisers.jl is supported by FluxTraining.jl, FastAI.jl should move to using them. This means - default optimisers come from Optimisers.jl - docs are updated - parameter groups and layer...

[https://github.com/libffcv/ffcv] seems to be the state-of-the-art when it comes to fast image pipelines for training deep learning models on computer vision. Would be interesting to benchmark against this and see...

help wanted
performance

Some low-hanging fruit that FastAI.jl can pick to become faster here: - builtin presizing support - prototype at https://github.com/lorenzoh/DataBlockStorage.jl/ - using JpegTurbo.jl's downsized loading so no storing to disk is...

enhancement
help wanted

Ideally, all the documentation pages based off of Jupyter notebooks should have a button to launch them in Colab. For example, fastai's documentation has such a button: ![image](https://user-images.githubusercontent.com/28812146/124114194-94a38680-da6c-11eb-9fb9-75dbccf9a860.png) First step...

documentation
help wanted

With #151, FastAI.jl is getting high-level interfaces for searching datasets (`finddatasets`) and loading datasets into task-specific data containers (`loaddataset`). There is also a new `DatasetRecipe` that encapsulates configuration for loading...

help wanted
good first issue

With the new Pollen.jl frontend having been adopted in https://github.com/FluxML/FastAI.jl/pull/203, I am taking the opportunity to think about changes and additions to the documentation. The term Reader refers to someone...

documentation
plans