Closed salelkafrawy closed 2 years ago
Hello Team, These are the required features to add for Q4 in milestone-2 document.
There were already existing features such as:
period
distance_from_net
angle
type
secondary_type
is_empty_net
coordinate_x
coordinate_y
And what has been done in this PR:
game_sec
Information from the immediately preceding event as four new features:
prev_event_type
prev_event_time_diff
distance_from_prev_event
Features to quantify a few more interesting things about the state of the play. Create the following three features:
is_rebound
rebound_angle
speed
What is still needs to be done:
tidy_data.py
events_df.head()
Hello Team, These are the required features to add for Q4 in milestone-2 document.
There were already existing features such as:
period
: at which period of the game the event was recorded in)distance_from_net
: distance between the shot and the net)angle
: the angle with respect to the net where the shot happened)type
andsecondary_type
: shot main and secondary types)is_empty_net
: a binary variable whether the net was empty or not)coordinate_x
,coordinate_y
, are the separate columns for the x and y coordinates respectively)And what has been done in this PR:
game_sec
: Total number of seconds elapsed in the game) 'time' columns is actually 'periodTime'Information from the immediately preceding event as four new features:
prev_event_type
: the type of the preceding event)prev_event_time_diff
: the difference between current event and the preceding event in seconds)distance_from_prev_event
: the distance between current event and the preceding one)Features to quantify a few more interesting things about the state of the play. Create the following three features:
is_rebound
: a binary column that is true if the current and preceding events were shots, otherwise false)rebound_angle
: only included if the shot is a rebound, otherwise 0. It is the angle between both shots).speed
: defined as the distance from the previous event, divided by the time since the previous event)What is still needs to be done:
tidy_data.py
script running timeevents_df.head()
and the previous description of the added columns to the blog post