制作一个 zip() pandas 列来总结具有唯一索引的其他列

Question

I have a DataFrame with 3 columns:

• store
• product
• price

For each store we have multiple products, but each product has a unique price. The DataFrame is hence composed of multiple rows on the same store, each row corresponding to a product.

I would like to make some transformations on the dataset to get only one line per store, and a compound column that would sum up info about products and prices as follow:

[(product_1,price_1),(product_2,price_2), ...]

For now I've not been able to do it.

What I have done is that I've grouped by store , aggregated by product, and applied the .unique() function. I get for each store, a list of all the products, but not the prices. When I try to add price to the .agg() function followed by .unique() it doesn't work and have no clue how to do this.

I guess I might have to apply some zipping at some point: zip(product, price) but I don't get until there.

Any help is appreciated, thanks!

Answer 1

df.groupby("store", as_index = False).apply(lambda x: pd.Series({'store': x["store"].iloc[0],
                                                                "result": [(val["product"], val["price"]) for idx, val in x.iterrows()]}))