The main goal behind this question is extracting the best practices on how to establish a Data Science / IA Project.
Anyone would be generous to share with us the set of tools or even a boilerplate that may help or guide anyone working in the field and applying data science & IA to any domain.
It very depend on dataset that are you going to used. Please specify the task tha are you going to solved. The preocess quite straight forward.
As Igor said, it depends on the dataset. However, you can follow some standard practices, such as using Kubeflow (a machine learning pipelining tool - https://www.kubeflow.org/docs/). Furthermore, below are some of the tools that you may find helpful.
Model serving:
- https://www.bentoml.ai
- https://www.cortex.dev
Production model monitoring -- to observe model degradation.
- https://github.com/evidentlyai/evidently
- https://www.seldon.io/tech/
- https://hydrosphere.io
- https://www.fiddler.ai/ml-monitoring
Note: Not all of them are open-source.
Dear Ihababdelbasset Annaki ,
I'd second what the others arleady said, it depends always a lot on the dataset and the task you want your data to perform. The pipelining approach also lets you easily forget about the power of simply playing around with your data. I for instance am primarily concerned with NLP where stemming and spelling correction are usually considered important steps in the data processing pipeline (and can be quickly implemented with frameworks such as spaCy, https://spacy.io/usage/processing-pipelines). I personally, however, usually end up with much better results, when I neglect them.
If you, however, want to model prototypic pipelines, let's say for didactic purposes, here are some nice introductions that also mention some tools associated with each step:
https://towardsdatascience.com/a-beginners-guide-to-the-data-science-pipeline-a4904b2d8ad3
https://dzone.com/articles/overview-of-the-data-science-pipeline
https://www.justintodata.com/data-science-pipeline-steps/
Additionally, if you still want to look into one-fits-all, or one-fits-almost-all solutions, have a look at Amazon's SageMaker for workflow organization:
https://www.analyticsvidhya.com/blog/2021/05/data-science-pipelines-and-workflow/
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts