wilsonrljr / sysidentpy

A Python Package For System Identification Using NARMAX Models
https://sysidentpy.org
BSD 3-Clause "New" or "Revised" License
373 stars 74 forks source link

Create M5 dataset example #100

Open wilsonrljr opened 1 year ago

wilsonrljr commented 1 year ago

One of the key features of SysIdentPy is its ability to handle time series data, making it a popular choice for many applications in various industries, including engineering and retail.

The SysIdentPy maintainer has set a goal to create a new example using the M5 competition dataset provided by Walmart. This dataset is interesting for several reasons, including its large scale and the fact that it features many time series with intermittency (sporadic sales including zeros).

By creating a new example using the M5 competition dataset, SysIdentPy will demonstrate its ability to handle large-scale time series data and address the challenges posed by sporadic demand. This new example will showcase the power of SysIdentPy in modeling and forecasting complex systems, such as those found in retail scenarios.

To ensure the success of this project, the SysIdentPy maintainer, wilsonrljr, will be providing expert guidance and support throughout the development process. This will include assisting with data preprocessing, providing guidance on model selection, and helping to optimize the performance of the resulting model.

The new example will be designed to be easy to use and accessible to a wide range of users, from beginners to advanced users. It will provide a step-by-step guide to building a predictive model using SysIdentPy, from data preprocessing to model evaluation.

In conclusion, the development of a new example using the M5 competition dataset is an important step forward for SysIdentPy and will demonstrate its ability to handle large-scale time series data and address the challenges posed by infrequent occurrences. With the support of the SysIdentPy maintainer, users can expect to have access to this new example in the near future. This will enable them to apply SysIdentPy to a wide range of real-world scenarios, including those found in retail and other industries.

PedroHPLopes commented 1 year ago

I would be happy to help with this If possible!!!

wilsonrljr commented 1 year ago

Hey @PedroHPLopes thanks! Can you send me a message on discord so we can talk to make a plan to work on this? You can find me by joining the SysIdentPy channel: https://discord.gg/8eGE3PQ