tangger/lerobot
1
0
Fork 1
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Hardcoded some normalization parameters. TODO refactor

Browse Source 
Added masking actions on the level of the intervention actions and offline dataset

Co-authored-by: Adil Zouitine <adilzouitinegm@gmail.com>
This commit is contained in:
Michel Aractingi
2025-02-13 14:27:14 +01:00
committed by AdilZouitine
parent a0e0a9a9b1
commit eb7e28d9d9
6 changed files with 36 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
lerobot/configs/policy/sac_real.yaml
View File

@@ -31,7 +31,7 @@ training:
online_env_seed: 10000
online_buffer_capacity: 1000000
online_buffer_seed_size: 0
online_step_before_learning: 1000 #5000
online_step_before_learning: 100 #5000
do_online_rollout_async: false
policy_update_freq: 1
@@ -61,7 +61,7 @@ policy:
observation.images.side: [3, 128, 128]
# observation.image: [3, 128, 128]
output_shapes:
action: ["${env.action_dim}"]
action: [4] # ["${env.action_dim}"]
# Normalization / Unnormalization
input_normalization_modes:
@@ -84,9 +84,12 @@ policy:
output_normalization_modes:
action: min_max
output_normalization_params:
# action:
# min: [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0]
# max: [1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
action:
min: [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0]
max: [1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
min: [-145.283203125, -69.43359375, -78.75, -46.0546875]
max: [145.283203125, 69.43359375, 78.75, 46.0546875]
# Architecture / modeling.
# Neural networks.