Changes in version 2.2.2 (2024-04-11)                  

  - Updated installation function install_fastai

                 Changes in version 2.2.1 (2023-03-11)                  

  - Fixed breaking changes during model training
  - Updated installation function install_fastai

                 Changes in version 2.0.9 (2021-07-28)                  

  - PyTorch version is 1.9
  - lr_find bug is fixed

                 Changes in version 2.0.8 (2021-05-16)                  

  - minor fixes

                 Changes in version 2.0.7 (2021-04-18)                  

  - bs_finder is fixed
  - better visualization on Colab. Issue with fig size is fixed

                 Changes in version 2.0.6 (2021-03-27)                  

  - new function to load_learner
  - unet_config is Deprecated
  - while installing fast.ai Mac OS, first, it downloads PyTorch 1.8,
    then 1.7.1. It is fixed, now.
  - nn_module() function allows to rename the model, e.g. summary(model)
  - nn_module() will not move the model to GPU, if gpu argument is FALSE
    (by default it is TRUE)
  - custom loss functions with nn_loss(). Based on Kaggle notebook

                 Changes in version 2.0.5 (2021-03-06)                  

  - install_fastai no more supports extensions. They need to be
    installed separately by users.
  - PyTorch was upgraded from 1.7.0 to 1.7.1.

                 Changes in version 2.0.4 (2021-01-29)                  

  - stick to fastaudio 0.1.3 (resolve dependencies)
  - add geom_point for interactive visualization within RStudio
  - add TPU module into fastai

                 Changes in version 2.0.3 (2021-01-09)                  

  - current stable version of fast.ai is 2.1.5
  - lots of new callback ops
  - freeze and unfreeze a model
  - object detection module - icevision
  - issue with exporting of a pickle file

                 Changes in version 2.0.2 (2020-12-09)                  

  - Hugging Face integration, prediction
  - one_batch() ability to add more arguments
  - no need to call options(reticulate.useImportHook = FALSE)
  - DataBlock automatically places data into cuda if available

                 Changes in version 2.0.1 (2020-11-12)                  

  - nn_module for model construction
  - fix_fit for disabling the training plot
  - all the fit functions now return the training history

                 Changes in version 2.0.0 (2020-11-11)                  

  - Initial release