You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
38 lines
1.3 KiB
38 lines
1.3 KiB
1. Objective:
|
|
We want to predict the vehicle type given other onboard metrics.
|
|
|
|
2. License: Free to use, collected by Splunk.
|
|
|
|
3. Data Source:
|
|
Splunk's annual Track Day!
|
|
|
|
4. Data Set Information:
|
|
The dataset contains 50,000 instantaneous measurements collected via onboard instrumentation from several vehicles while being driven around a racetrack.
|
|
|
|
5. Field Meanings:
|
|
A. batteryVoltage - battery voltage in volts (surprise!)
|
|
B. engineCoolantTemperature - degrees celcius
|
|
C. engineSpeed - rotations per minute
|
|
D. lateralGForce - left/right
|
|
E. longitudeGForce - front/back
|
|
F. speed - miles per hour
|
|
G. vehicleType - year, make, and model
|
|
H. verticalGForce - up/down
|
|
|
|
6. Parameter Selection:
|
|
A. Dashboard: Predict Categorical Fields
|
|
Settings:
|
|
1) Field to predict: vehicleType
|
|
2) Fields to use for predicting: batteryVoltage, engineCoolantTemperature, engineSpeed, lateralGForce, longitudeGForce, speed, and verticalGForce
|
|
|
|
B. Dashboard: Cluster Numeric Fields
|
|
Settings:
|
|
0) Search: | inputlookup track_day.csv
|
|
1) Model name: track_day
|
|
2) Fields to preprocess: batteryVoltage, engineCoolantTemperature, engineSpeed, lateralGForce, longitudeGForce, speed, verticalGForce
|
|
3) Apply StandardScaler
|
|
4) Apply PCA to reduce to 3 fields
|
|
5) Algorithm: Birch
|
|
6) Fields to use: PC_1, PC_2, PC_3
|
|
7) K: 6
|