Armaghan Shakir

Results 17 issues of Armaghan Shakir

This issue tracks the change cycle for renaming - `cINNForecaster` to `CINNForecaster` in `sktime.forecasting.conditional_invertible_neural_network` - `cINNNetwork` to `CINNNetwork` in `sktime.networks.cinn` The change cycle will be implemented in 3 steps with...

maintenance
module:forecasting

#### Reference Issues/PRs This completes step 3 of change cycle for renaming cINNForecaster to CINNForecaster, completing the steps for release v0.30.0 in #6120 #### What does this implement/fix? Explain your...

maintenance
release

To extend PEFT fine tuning methods in the existing huggingface interface, we need to wrap the model initialization with peft configuration. Hypothetically, this can be done using this piece of...

module:forecasting
enhancement

This pull request adds a new example demonstrating the integration of Qdrant Search Engine into various applications. The example provides a step-by-step guide for building a search engine very quickly...

Some examples are not listed in the markdown table. Is there a specific reason to keep them away? Because the visitor is most probably going to navigate through the links...

**Mentee:** Armaghan Shakir (he), [Linkedin](https://www.linkedin.com/in/armaghan-shakir/) **Mentor(s):** @fkiraly @benHeid @yarnabrina **Why did you join sktime's mentorship program?** For a long time, I've wanted to contribute and collaborate on open source projects....

mentoring

#### Reference Issues/PRs Implements `LTSFTransformer` from https://github.com/sktime/sktime/issues/4939 #### What does this implement/fix? Explain your changes. New forecaster `LTSFTransformer` #### Does your contribution introduce a new dependency? If yes, which one?...

module:forecasting
enhancement

#### Reference Issues/PRs Fixes #6435. #### What does this implement/fix? Explain your changes. This PR extends the `fit_strategy` of `HFTransformersForecaster` for these `PEFT` methods 1. LoRa 2. LoHa 3. AdaLora...

module:forecasting
enhancement

`cINNForecaster` interface currently needs some error handling and documentation improvements as discussed here: https://github.com/sktime/sktime/issues/6122 In `cINNForecaster`, it is important to select the hyper parameters carefully. The default curve fit tries...

good first issue
module:forecasting
enhancement

Update the docstring for parameter `model_path` to more informative info.