Mogptk

- Added sampling to all likelihoods
- [Add exact variational expectation implementation for exponential likelihood](https://github.com/GAMES-UChile/mogptk/commit/9d5347937196e621274e67e8338c238f466446ac)
- [Add exact variational expectation implementation for Poisson likelihood](https://github.com/GAMES-UChile/mogptk/commit/f3b9883662cf82c9dfa73bb9cbdd806fa8b0e619)
- [Add exact variational expectation implementation for Gamma likelihood](https://github.com/GAMES-UChile/mogptk/commit/0c6134a0ae2ab56f6ca98942b4a81c16b75b0ee0)
- [Add sampling y values](https://github.com/GAMES-UChile/mogptk/commit/d82a67257119f6eca96b23c157fa0e6d95f4bf63)
- Bug fixes in likelihoods

Added mathematical notes to repository at: https://github.com/GAMES-UChile/mogptk/blob/master/notes/notes.pdf

- Many many bug fixes
- Revise all examples in the documentation
- Revert to default `float64` dtype instead of `float32` to avoid precision errors
- Improve verbose training output
- Improve plots slightly
- Add `mogptk.gpr.MultiOutputMean`, support for different mean functions for each output dimension
- Make all randomness come from PyTorch and not also numpy
- Revert memory optimization for exact model to avoid Cholesky problems
- **[Add prediction confidence intervals and likelihood sampling](https://github.com/GAMES-UChile/mogptk/commit/665e3d385ef6de5d22e7b3e9e7a66449db667ef0)**

- [Use `inference_mode` instead of `no_grad`, slight increase in performance](https://github.com/GAMES-UChile/mogptk/commit/20c0146092755af56dc8971930c7e4fb0ec94e04)
- [Big change: all classes now inherit from `torch.nn.Module` or `torch.nn.Parameter`; refactoring of parameters; removal of naming for kernels/means/likelihoods](https://github.com/GAMES-UChile/mogptk/commit/e3c01d7bdb1219f538d7554dc81bf91367d70d70)
- [Rename copy_parameters to load_kernel_parameters; get_parameters to parameters; add __str__ method to model](https://github.com/GAMES-UChile/mogptk/commit/5f7a229c39476ef2425f2aaa52b401d490313f2b)
- [Optimization: set grad to None instead of zero](https://github.com/GAMES-UChile/mogptk/commit/6e9c31d437d2c1bff0fb642d98061b4c7ae42c0b)
- [Optimization: reduce memory of Exact model by not preallocating the 'eye' tensor for the Gram matrix](https://github.com/GAMES-UChile/mogptk/commit/df308ed9da565a1016dffc472db947998c862811)
- **[Support `torch.jit` by default for training, improves performance by about 10% after the first two iterations (warmup)](https://github.com/GAMES-UChile/mogptk/commit/681b7b5ae60635cd277568c4be44c4e523cb1f3b)**

- [Improve accuracy of time interval to display training progress](https://github.com/GAMES-UChile/mogptk/commit/51db99897e6cd6a5fa6802380ba7e1ee640f1365)
- [Add error when prediction data is of wrong shape](https://github.com/GAMES-UChile/mogptk/commit/b6e41fe81a9e069103edff09931b845ad1d35b44)
- [Accept torch.Tensor for Data and DataSet](https://github.com/GAMES-UChile/mogptk/commit/5fab287ca0b1417e364e178c29274ee1eab3518a)
- [Bugfixes and 10% speed improvement due to caching parameter transformation](https://github.com/GAMES-UChile/mogptk/commit/ffda2673522bbb262da689dc03dec417488d73b4)
- Default to float32 instead of float64 for PyTorch, this will reduce memory usage by default. Use `mogptk.gpr.use_double_precision()` to revert to the previous default.

- [Fix error when using the non-sparse Hensman model](https://github.com/GAMES-UChile/mogptk/commit/ae777b90752156fccdebbbca47263d68264ef6f1)
- [Fix a bug for use of undefined device and dtype in Kernel](https://github.com/GAMES-UChile/mogptk/commit/62ad8421d3f9d72a0f6f536e909b8e34d1c5dffd)
- [Fix Snelson and OpperArchambeau models,](https://github.com/GAMES-UChile/mogptk/commit/8a6be439b4fc62913718adcb7c51436a258c83ee) [fixes](https://github.com/GAMES-UChile/mogptk/commit/8a6be439b4fc62913718adcb7c51436a258c83ee) https://github.com/GAMES-UChile/mogptk/issues/58 [and](https://github.com/GAMES-UChile/mogptk/commit/8a6be439b4fc62913718adcb7c51436a258c83ee) [fixes](https://github.com/GAMES-UChile/mogptk/commit/8a6be439b4fc62913718adcb7c51436a258c83ee) https://github.com/GAMES-UChile/mogptk/issues/59
- [Fix inplace operation warning for PyTorch v2.0.0,](https://github.com/GAMES-UChile/mogptk/commit/b9509399a77162cf6764f41c45161576064e39df) [fixes](https://github.com/GAMES-UChile/mogptk/commit/b9509399a77162cf6764f41c45161576064e39df) https://github.com/GAMES-UChile/mogptk/issues/53

- [Fix bug in `LinearMean` for multiple input dimensions](https://github.com/GAMES-UChile/mogptk/commit/187f1887b969c8ad66103148c8f0005dd29e1678)
- [Rename inference model parameter from `model` to `inference`](https://github.com/GAMES-UChile/mogptk/commit/c4bfff1375a348c212618334e3ccb413bc873e0d)
- [Fix bug in `copy_parameters`](https://github.com/GAMES-UChile/mogptk/commit/c4bfff1375a348c212618334e3ccb413bc873e0d)