how to get the dataset interactions.csv

[gccrpm]

hi , @mquad ,i can't get dataset from XING.com ,could you tell how to get the dataset?
thank you!

[mquad]

Hi @gccrpm , sadly the dataset has been removed by the owners (see #1 ). You can try out other public datasets such as the Retailrocket one (https://www.kaggle.com/retailrocket/ecommerce-dataset)

[simushga]

Hi,

I tried to run "python build_dataset.py interactions.csv". It says:

Loading interactions.csv
Building sessions
Original data:
Num items: 2314
Num users: 1819
Num sessions: 3189
Filtering data
Filtered data:
Num items: 0
Num users: 0
Num sessions: 0
Partitioning data
Write to disk

I appreciate if you help me understand why filtered data is all 0.

Here is how my data looks like:

user_id item_id interaction_type created_at
100001 214004541 1 1515827044
100002 214006968 1 1523192543
100003 214005492 1 1515076970

Thanks
Sima

[simushga]

I found the answer to my above question. Filtering was too tight for my sample mini dataset. Namely, these two were removing everything:

keep items with >=20 interactions
let's keep only returning users (with >= 5 sessions)

Thanks

[mquad]

Happy to see that you found the solution 😃

[chordou]

Hi,

I tried to run "python build_dataset.py interactions.csv". It says:

Loading interactions.csv
Building sessions
Original data:
Num items: 2314
Num users: 1819
Num sessions: 3189
Filtering data
Filtered data:
Num items: 0
Num users: 0
Num sessions: 0
Partitioning data
Write to disk

I appreciate if you help me understand why filtered data is all 0.

Here is how my data looks like:

user_id item_id interaction_type created_at
100001 214004541 1 1515827044
100002 214006968 1 1523192543
100003 214005492 1 1515076970

Thanks
Sima

sorry to bother you. Would you mind providing me dataset? Thank you for your help!