Datasets:
Create README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,36 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
The ID column links tables in X_train, with Y_train and Y_train_supp. The same holds true for the test data.
|
| 2 |
+
|
| 3 |
+
Input team data sets comprise the following 3 identifier columns:
|
| 4 |
+
|
| 5 |
+
ID, LEAGUE and TEAM_NAME (note that LEAGUE and TEAM_NAME are not included in the test data)
|
| 6 |
+
The following 25 statistics, which are aggregated by sum, average and standard deviation.
|
| 7 |
+
|
| 8 |
+
'TEAM_ATTACKS'
|
| 9 |
+
'TEAM_BALL_POSSESSION'
|
| 10 |
+
'TEAM_BALL_SAFE'
|
| 11 |
+
'TEAM_CORNERS'
|
| 12 |
+
'TEAM_DANGEROUS_ATTACKS'
|
| 13 |
+
'TEAM_FOULS'
|
| 14 |
+
'TEAM_GAME_DRAW'
|
| 15 |
+
'TEAM_GAME_LOST'
|
| 16 |
+
'TEAM_GAME_WON'
|
| 17 |
+
'TEAM_GOALS'
|
| 18 |
+
'TEAM_INJURIES'
|
| 19 |
+
'TEAM_OFFSIDES'
|
| 20 |
+
'TEAM_PASSES'
|
| 21 |
+
'TEAM_PENALTIES'
|
| 22 |
+
'TEAM_REDCARDS'
|
| 23 |
+
'TEAM_SAVES'
|
| 24 |
+
'TEAM_SHOTS_INSIDEBOX'
|
| 25 |
+
'TEAM_SHOTS_OFF_TARGET'
|
| 26 |
+
'TEAM_SHOTS_ON_TARGET',
|
| 27 |
+
'TEAM_SHOTS_OUTSIDEBOX'
|
| 28 |
+
'TEAM_SHOTS_TOTAL'
|
| 29 |
+
'TEAM_SUBSTITUTIONS'
|
| 30 |
+
'TEAM_SUCCESSFUL_PASSES'
|
| 31 |
+
'TEAM_SUCCESSFUL_PASSES_PERCENTAGE'
|
| 32 |
+
'TEAM_YELLOWCARDS'
|
| 33 |
+
Input player data sets comprise the following 3 identifier columns:
|
| 34 |
+
|
| 35 |
+
ID, LEAGUE and TEAM_NAME, POSITION and PLAYER_NAME (note that LEAGUE, TEAM_NAME, and PLAYER_NAME are not included in the test data)
|
| 36 |
+
52 statistics, which are aggregated by sum, average and standard deviation. They are similar to the team statistics though more fine-grained.
|