gouhanke
38 lines
681 B
YAML
38 lines
681 B
YAML
# @package agent
|
|
defaults:
|
|
- /backbone@vision_backbone: resnet_diffusion
|
|
- /modules@state_encoder: identity_state_encoder
|
|
- /modules@action_encoder: identity_action_encoder
|
|
- /head: gr00t_dit1d
|
|
- _self_
|
|
|
|
_target_: roboimi.vla.agent_gr00t_dit.VLAAgentGr00tDiT
|
|
|
|
# Model dimensions
|
|
action_dim: 16
|
|
obs_dim: 16
|
|
|
|
# Normalization
|
|
normalization_type: "min_max"
|
|
|
|
# Horizons
|
|
pred_horizon: 16
|
|
obs_horizon: 2
|
|
num_action_steps: 8
|
|
|
|
# Cameras
|
|
num_cams: 3
|
|
|
|
# Diffusion
|
|
diffusion_steps: 100
|
|
inference_steps: 10
|
|
|
|
# Head overrides
|
|
head:
|
|
input_dim: ${agent.action_dim}
|
|
output_dim: ${agent.action_dim}
|
|
horizon: ${agent.pred_horizon}
|
|
n_obs_steps: ${agent.obs_horizon}
|
|
cond_dim: 208
|
|
|