Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] Blog about working with Time Series Data using FastAI.jl #140

Open
wants to merge 4 commits into
base: main
Choose a base branch
from
Open
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
resolving comments
  • Loading branch information
codeboy5 committed Sep 18, 2022
commit 4eb6316085d9563df044b1f419c7bb1ef8945e5a
23 changes: 20 additions & 3 deletions blog/_posts/2022-09-08-Adding-Time-Series-Support-to-FastAI.jl.md
Original file line number Diff line number Diff line change
Expand Up @@ -49,13 +49,26 @@ julia> encodesample(task, Training(), (input, class))
## Models

The library contains implementation of the following models.
- Basic stacked RNNs

- Basic stacked RNNs.
Stacked RNN network. Feeds the data through a chain of RNN layers, where the hidden state
of the previous layer gets fed to the next one. The Model has the following arguments.
- `c_in` : The number of input channels.
- `c_out` : The number of output classes.
- `hiddensize` : The number of "hidden channels" to use.
- `layers` : The number of RNN layers to use in the stacked network.

```julia
julia> backbone = FastTimeSeries.Models.StackedLSTM(1, 16, 10, 2);
julia> model = FastAI.taskmodel(task, backbone);
```

- [InceptionTime](https://arxiv.org/abs/1909.04939)
- [InceptionTime](https://arxiv.org/abs/1909.04939)
An implementation of the InceptionTime Model. The Model has the following arguments.
- `c_in` : The number of input channels.
- `c_out` : The number of output classes.
- `nf` : The number of "hidden channels" to use.

```julia
julia> model = FastTimeSeries.Models.InceptionTime(1, 5);
```
Expand All @@ -78,7 +91,11 @@ We can view the loss and accuracy on the training and validation data after the

## Conclusion

We saw how we can work on time-series data using FastAI.jl.
We saw how we could work on time-series data using FastAI.jl. The library now supports the following features.

- Load time series classification datasets and regression datasets from [UCR Classification](http://timeseriesclassification.com/index.php) and [Monash Regression](http://tseregression.org/) respectively.
- Perform data transformations on the loaded data.
- Perform classification and regression tasks using InceptionTime or stacked RNNs.

## Future Work

Expand Down