I've got product data (
time, title, price, likes, colour, etc.) from several shops and I would like to do some analysis on it. Some are quite easy:
- Total amount of products per shop
- Pie Chart of colour distribution per country
- Amount of likes over time (line chart)
- Price changes
What I'm currently using is Python + Pandas for most of the transformation/data manipulation etc. From there I convert DataFrames into dicts and dump them as JSON files. Those are being loaded via Chart.js.
That is working ok but it takes me hours, if not days for every new idea of what kind of analisys I'd like to do or which answer I'd like to visualise.
Then I read about TensorFlow and it all sounds so awesome - maybe that would be a way to go to analyse trends or get more knowledge out of my data?
Or would a traditional BI do a sufficient job here?
Maybe just JasperSoft to visualise the data?
I don't have a specific question/analytics requirement - it's more like a hobby project which is growing over time.