# **meeting 01/02** **Advisor: Prof. Chih-Yu Wang \ Presenter: Shao-Heng Chen \ Date: Jan 02, 2023** <!-- Chih-Yu Wang --> <!-- Wei-Ho Chung --> ## **Inference result** #### ```PPO-2-16-4``` <!-- default: 3-0 --> ```shell -------------------------------- random action: mean: -11.590974656626582 std: 9.071662609528438 max: -0.2852657437324524 min: -48.524688720703125 shape: (220,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-4-mse: mean: -1.7125871181488037 std: 1.2584214210510254 max: -0.017516791820526123 min: -8.086286544799805 shape: (1, 220) -------------------------------- ``` <img src='https://hackmd.io/_uploads/B1gAuYSUua.png' width=50% weight=50%> <img src='https://hackmd.io/_uploads/Bke0dFBUOp.png' width=50% weight=50%> #### ```PPO-2-16-16``` <!-- 0-1--> ```shell -------------------------------- random action: mean: -42.11586773693561 std: 30.182951263098346 max: -0.8216069936752319 min: -214.58811950683594 shape: (688,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-16-mse: mean: -10.529911994934082 std: 7.090195655822754 max: -0.7868634462356567 min: -60.0086784362793 shape: (1, 688) -------------------------------- ``` <img src='https://hackmd.io/_uploads/S162OBL_6.png' width=50% weight=50%> <img src='https://hackmd.io/_uploads/rJThOHUup.png' width=50% weight=50%> <img src='' width=50% weight=50%> <img src='' width=50% weight=50%> <!-- ```186_0000``` ```shell -------------------------------- random action: mean: -42.187725625157356 std: 28.85515651599719 max: -1.2918919324874878 min: -198.29696655273438 shape: (688,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-16-mse: mean: -9.23713493347168 std: 6.263749122619629 max: -0.6578090190887451 min: -45.7191276550293 shape: (1, 688) -------------------------------- ``` ```118_0000``` ```shell -------------------------------- random action: mean: -42.32311049896479 std: 29.511914605443827 max: -1.2089446783065796 min: -217.38758850097656 shape: (688,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-16-mse: mean: -9.948446273803711 std: 6.551253318786621 max: -0.783655047416687 min: -50.91864776611328 shape: (1, 688) -------------------------------- ``` --> #### ```PPO-2-16-36``` <!-- 0-0 --> ```shell -------------------------------- random action: mean: -66.2713088144064 std: 48.67355126857339 max: -1.6711686849594116 min: -352.4002380371094 shape: (1468,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-36-mse: mean: -47.03792953491211 std: 34.761619567871094 max: -0.8887106776237488 min: -256.8116149902344 shape: (1, 1468) -------------------------------- ``` <img src='https://hackmd.io/_uploads/rkPcqH3PT.png' width=50% weight=50%> <img src='https://hackmd.io/_uploads/Sys9crhvT.png' width=50% weight=50%> <!-- ```156_0000``` ```shell -------------------------------- random action: mean: -63.80303210687637 std: 45.61789485282826 max: -1.4327713251113892 min: -311.0680236816406 shape: (1468,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-36-mse: mean: -47.696449279785156 std: 34.956790924072266 max: -1.5138236284255981 min: -248.5370330810547 shape: (1, 1468) -------------------------------- ``` ```110_0000``` ```shell -------------------------------- random action: mean: -65.00288468801975 std: 47.89249963272243 max: -1.1998952627182007 min: -354.20086669921875 shape: (1468,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-36-mse: mean: -57.305484771728516 std: 39.498844146728516 max: -1.3809412717819214 min: -295.75115966796875 shape: (1, 1468) -------------------------------- ``` --> #### ```PPO-2-16-64``` <!-- default 0-0 --> <!-- ```200_0000``` ```shell -------------------------------- random action: mean: -105.64727407515049 std: 79.5604244250156 max: -1.9135271310806274 min: -677.9887084960938 shape: (2560,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-64-mse: mean: -47.507415771484375 std: 34.10666275024414 max: -0.4617055058479309 min: -235.93228149414062 shape: (1, 2560) -------------------------------- ``` --> ```shell -------------------------------- random action: mean: -103.00651884531975 std: 76.50868783408355 max: -3.1422348022460938 min: -575.050537109375 shape: (2560,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-64-mse: mean: -50.01382827758789 std: 36.21616744995117 max: -1.0844701528549194 min: -257.2842712402344 shape: (1, 2560) -------------------------------- ``` <img src='https://hackmd.io/_uploads/H1vsjSnvp.png' width=50% weight=50%> <img src='https://hackmd.io/_uploads/SJUosBhvT.png' width=50% weight=50%> <!-- ```18_0000``` ```shell -------------------------------- random action: mean: -104.74705203723907 std: 76.71905901209834 max: -1.9807257652282715 min: -613.1129150390625 shape: (2560,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-64-mse: mean: -44.69156265258789 std: 31.410484313964844 max: -0.834943413734436 min: -246.3562774658203 shape: (1, 2560) -------------------------------- ``` --> #### ```PPO-2-16-100``` <!-- default 0-0 --> <!-- ```180_000``` ```shell -------------------------------- random action: mean: -124.07158871620894 std: 92.79494145479518 max: -1.9796050786972046 min: -681.3377685546875 shape: (3964,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-100-mse: mean: -0.8677553534507751 std: 0.3841693103313446 max: -0.16700375080108643 min: -3.379624366760254 shape: (1, 3964) -------------------------------- ``` --> <!-- 0-1 --> ```200_000``` ```shell -------------------------------- model inference of PPO-2023-12-27-2-16-100-mse: mean: -119.98394725298881 std: 87.72033508359831 max: -2.3737082481384277 min: -635.4803466796875 shape: (3964,) -------------------------------- -------------------------------- model inference of PPO-2023-12-27-2-16-100-mse: mean: -51.55144500732422 std: 37.97480773925781 max: -1.011149525642395 min: -277.65948486328125 shape: (1, 3964) -------------------------------- ``` <img src='https://hackmd.io/_uploads/rkJvULedT.png' width=50% weight=50%> <img src='https://hackmd.io/_uploads/BJkP8Uldp.png' width=50% weight=50%> ## **Confidence Interval** <img src='https://hackmd.io/_uploads/rkHtZvl_p.png' width=55% weight=50%> <img src='https://hackmd.io/_uploads/SJTFxDg_p.png' width=50% weight=50%> <img src='https://hackmd.io/_uploads/ryqpeDedT.png' width=50% weight=50%> <img src='https://hackmd.io/_uploads/HkN3ePg_a.png' width=50% weight=50%> <img src='https://hackmd.io/_uploads/S1c2ePeup.png' width=50% weight=50%>