Open PaParaZz1 opened 3 years ago
Evogym Carrier-v0 random bot VS trained agent
agent during early training VS trained agent
Procgen Bigfish random bot VS trained agent
Metadrive env with PPO algorithm demonstration. Map Sequence: Straight + Intersection + Straight + Roundabout
cliffwalking agent trained by DQN random agent vs trained agent
https://github.com/opendilab/DI-engine/assets/93968541/756cbba6-4405-4882-8753-6e4288e40a40
Hi @zjowowen, thank you for your sharing. I would like to ask about fly through gate simulation, did you change the compute reward and terminal function or you use the same with original work.
This issue is a collection of various interesting agent demonstration trained by DI-engine, it will be updated continually.
Mario 1-1
https://user-images.githubusercontent.com/33195032/184529481-f55642f7-e9e8-45d3-9b8c-80e5bf3e177b.mp4
Mario 1-2
https://user-images.githubusercontent.com/33195032/184669648-6309ba32-4d33-4b07-8ce0-cce89c82bf97.mp4
rocket landing
https://user-images.githubusercontent.com/33195032/204750411-2c3c4fb3-de6c-474a-a4d3-dd380a605de1.mp4
SMAC 5m VS 6m
https://user-images.githubusercontent.com/33195032/125911080-2fdbd8ce-1107-455b-9772-eac20b717dab.mp4
SMAC MMM
https://user-images.githubusercontent.com/33195032/125910360-de914769-f0fc-4073-85a7-b12b1ef32fa7.mp4
SMAC MMM2
https://user-images.githubusercontent.com/33195032/126481161-9127f790-bfca-4985-bed3-7adaab519b46.mp4
SMAC 3s5z
https://user-images.githubusercontent.com/33195032/125908605-d694bb6a-b10a-4b58-9ec2-b4eb92cfe7a3.mp4
lunarlander
https://user-images.githubusercontent.com/33195032/127827210-cbc1e23c-6ed4-4d5b-a3c4-34a010119993.mp4
gfootball
https://user-images.githubusercontent.com/33195032/125933672-db8a6229-c2ea-4ab6-a53e-355a2258a0f3.mp4
https://user-images.githubusercontent.com/33195032/125933646-e24df3ec-9ec3-41d7-a693-136331f6a27a.mp4
slime_volley