dc8e0c9edc
* Sampling over both env and denoising steps in DPPO updates (#13) * sample one from each chain * full random sampling * Add Proficient Human (PH) Configs and Pipeline (#16) * fix missing cfg * add ph config * fix how terminated flags are added to buffer in ibrl * add ph config * offline calql for 1M gradient updates * bug fix: number of calql online gradient steps is the number of new transitions collected * add sample config for DPPO with ta=1 * Sampling over both env and denoising steps in DPPO updates (#13) * sample one from each chain * full random sampling * fix diffusion loss when predicting initial noise * fix dppo inds * fix typo * remove print statement --------- Co-authored-by: Justin M. Lidard <jlidard@neuronic.cs.princeton.edu> Co-authored-by: allenzren <allen.ren@princeton.edu> * update robomimic configs * better calql formulation * optimize calql and ibrl training * optimize data transfer in ppo agents * add kitchen configs * re-organize config folders, rerun calql and rlpd * add scratch gym locomotion configs * add kitchen installation dependencies * use truncated for termination in furniture env * update furniture and gym configs * update README and dependencies with kitchen * add url for new data and checkpoints * update demo RL configs * update batch sizes for furniture unet configs * raise error about dropout in residual mlp * fix observation bug in bc loss --------- Co-authored-by: Justin Lidard <60638575+jlidard@users.noreply.github.com> Co-authored-by: Justin M. Lidard <jlidard@neuronic.cs.princeton.edu>
59 lines
1.3 KiB
YAML
59 lines
1.3 KiB
YAML
defaults:
|
|
- _self_
|
|
hydra:
|
|
run:
|
|
dir: ${logdir}
|
|
_target_: agent.pretrain.train_gaussian_agent.TrainGaussianAgent
|
|
|
|
name: ${env}_pre_gaussian_mlp_ta${horizon_steps}
|
|
logdir: ${oc.env:DPPO_LOG_DIR}/gym-pretrain/${name}/${now:%Y-%m-%d}_${now:%H-%M-%S}_${seed}
|
|
train_dataset_path: ${oc.env:DPPO_DATA_DIR}/gym/${env}/train.npz
|
|
|
|
seed: 42
|
|
device: cuda:0
|
|
env: kitchen-partial-v0
|
|
obs_dim: 60
|
|
action_dim: 9
|
|
horizon_steps: 4
|
|
cond_steps: 1
|
|
|
|
wandb:
|
|
entity: ${oc.env:DPPO_WANDB_ENTITY}
|
|
project: gym-${env}-pretrain
|
|
run: ${now:%H-%M-%S}_${name}
|
|
|
|
train:
|
|
n_epochs: 5000
|
|
batch_size: 128
|
|
learning_rate: 1e-3
|
|
weight_decay: 1e-6
|
|
lr_scheduler:
|
|
first_cycle_steps: 5000
|
|
warmup_steps: 1
|
|
min_lr: 1e-4
|
|
epoch_start_ema: 10
|
|
update_ema_freq: 5
|
|
save_model_freq: 1000
|
|
|
|
model:
|
|
_target_: model.common.gaussian.GaussianModel
|
|
network:
|
|
_target_: model.common.mlp_gaussian.Gaussian_MLP
|
|
mlp_dims: [256, 256, 256]
|
|
activation_type: Mish
|
|
fixed_std: 0.1
|
|
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
|
|
horizon_steps: ${horizon_steps}
|
|
action_dim: ${action_dim}
|
|
horizon_steps: ${horizon_steps}
|
|
device: ${device}
|
|
|
|
ema:
|
|
decay: 0.995
|
|
|
|
train_dataset:
|
|
_target_: agent.dataset.sequence.StitchedSequenceDataset
|
|
dataset_path: ${train_dataset_path}
|
|
horizon_steps: ${horizon_steps}
|
|
cond_steps: ${cond_steps}
|
|
device: ${device} |