MovieLens Datasets #147
MovieLens Datasets #147
Conversation
Dsantra92
commented
Jun 23, 2022
•
Dsantra92
commented
- MovieLens Struct
- Base.show
- Docs
- MovieLens 100K
- MovieLens 1m
- MoviLens 20m
- MovieLens 25m
- Tests
Codecov Report
@@ Coverage Diff @@
## master #147 +/- ##
==========================================
+ Coverage 38.68% 44.01% +5.32%
==========================================
Files 39 40 +1
Lines 1755 2029 +274
==========================================
+ Hits 679 893 +214
- Misses 1076 1136 +60
Continue to review full report at Codecov.
|
src/datasets/graphs/movielens.jl
Outdated
| user_data["gender"] = user_df[!, 3] .== "M" # I hope I don't get cancelled for binarizing this field | ||
| user_data["occupation"] = user_df[!, 4] | ||
| user_data["zipcode"] = user_df[!, 5] | ||
| return user_data |
There was a problem hiding this comment.
The indentation is not uniform in this file. It should be 4 blanks everywhere
There was a problem hiding this comment.
Yeah, vim messed it up somewhere, I will fix these things in a cleanup commit later.
|
Forgot to add the indentation fix change, will be added in later commit. |
|
There are inconsistencies in data storing format across the 3 variations: 100k,1m and current datasets(20m, 25m etc.). Will address the issue when all of them have working APIs. |
|
Are all tests passing locally? |
Yes |
