sartajbhuvaji/self-driving-GTA-V

Self Driving GTA V Dataset

Dataset Varients

• Mini : Link
• Training Data(1-100) : Link
• Training Data(101-200) : Link

Info

• Image Resolution : 270, 480
• Mode : RGB
• Dimension : (270, 480, 3)
• File Count : 100
• Size : 1.81 GB/file
• Total Data Size : 362 GB
• Total Frames : 1 Million

Data Set sizes

Mini :

• Folder Name : mini
• Files : 01
• Total Size : 1.81 GB
• Total Frames : 5000

First Half

• Folder Name : Training Data(1-100)
• Files : 100
• Total Size : 181 GB
• Total Frames : 500,000

Second Half

• Folder Name : Training Data(101-200)
• Files : 100
• Total Size : 181 GB
• Total Frames : 500,000

Data Count

Mini

 'W':  [1, 0, 0, 0, 0, 0, 0, 0, 0] : 3627
 'S':  [0, 1, 0, 0, 0, 0, 0, 0, 0] : 50
 'A':  [0, 0, 1, 0, 0, 0, 0, 0, 0] : 104
 'D':  [0, 0, 0, 1, 0, 0, 0, 0, 0] : 106
 'WA': [0, 0, 0, 0, 1, 0, 0, 0, 0] : 364
 'WD': [0, 0, 0, 0, 0, 1, 0, 0, 0] : 416
 'SA': [0, 0, 0, 0, 0, 0, 1, 0, 0] : 35
 'SD': [0, 0, 0, 0, 0, 0, 0, 1, 0] : 47
 'NK': [0, 0, 0, 0, 0, 0, 0, 0, 1] : 248
  NONE : 3 

First Half (Data Count (1-100))

 'W':  [1, 0, 0, 0, 0, 0, 0, 0, 0] : 353725
 'S':  [0, 1, 0, 0, 0, 0, 0, 0, 0] : 2243
 'A':  [0, 0, 1, 0, 0, 0, 0, 0, 0] : 14303
 'D':  [0, 0, 0, 1, 0, 0, 0, 0, 0] : 13114
 'WA': [0, 0, 0, 0, 1, 0, 0, 0, 0] : 30877
 'WD': [0, 0, 0, 0, 0, 1, 0, 0, 0] : 29837
 'SA': [0, 0, 0, 0, 0, 0, 1, 0, 0] : 1952
 'SD': [0, 0, 0, 0, 0, 0, 0, 1, 0] : 1451
 'NK': [0, 0, 0, 0, 0, 0, 0, 0, 1] : 52256
  NONE : 242

Second Half (Data Count (101-200))

 'W':  [1, 0, 0, 0, 0, 0, 0, 0, 0] : 359025
 'S':  [0, 1, 0, 0, 0, 0, 0, 0, 0] : 2834
 'A':  [0, 0, 1, 0, 0, 0, 0, 0, 0] : 11025
 'D':  [0, 0, 0, 1, 0, 0, 0, 0, 0] : 9639
 'WA': [0, 0, 0, 0, 1, 0, 0, 0, 0] : 31896
 'WD': [0, 0, 0, 0, 0, 1, 0, 0, 0] : 29756
 'SA': [0, 0, 0, 0, 0, 0, 1, 0, 0] : 1742
 'SD': [0, 0, 0, 0, 0, 0, 0, 1, 0] : 2461
 'NK': [0, 0, 0, 0, 0, 0, 0, 0, 1] : 51313
  NONE : 309

Graphics Details

• Original Resolution : 800 x 600
• Aspect Ratio : 16:10
• All Video Settings : Low

Camera Details

• Camera : Hood Cam
• Vehical Camera Height : Low
• First Person Vehical Auto-Center : On
• First Person Head Bobbing : Off

Other Details

• Vehical : Michael's Car
• Vehical Mods : All Max
• Cv2 Mask : None
• Way Point : Enabled/Following
• Weather Conditions : Mostly Sunny
• Time of Day : Day, Night
• Rain : Some

Note

• Remove NONE while processing the data
• Use the mini dataset for initial setup and testing
• Check training_data_count_001-100.csv & training_data_count_101-200.csv for detailed count
• Check training_data_stats.py for more info

Inspired From

• Sentdex
• Youtube: Python Plays: Grand Theft Auto V