diff --git a/configs.yaml b/configs.yaml index 817d62d..82f14da 100644 --- a/configs.yaml +++ b/configs.yaml @@ -149,30 +149,33 @@ craftax: imag_gradient: 'reinforce' time_limit: 4000 -# craftax_small: -# task: craftax_Craftax-Symbolic-AutoReset-v1 -# step: 1e6 -# action_repeat: 1 -# envs: 1 -# train_ratio: 512 -# video_pred_log: false -# dyn_hidden: 256 -# dyn_deter: 256 -# dyn_stoch: 24 -# dyn_discrete: 24 -# # note: depth is cnn hidden_dim -# encoder: {cnn_keys: 'state_map', cnn_depth: 32, kernel_size: 4, minres: 2, mlp_keys: "state_inventory", mlp_layers: 2, mlp_units: 16} -# decoder: {cnn_keys: 'state_map', cnn_depth: 32, kernel_size: 4, minres: 2, mlp_keys: "state_inventory", mlp_layers: 2, mlp_units: 16} -# actor: {layers: 3, dist: 'onehot', std: 'none'} -# value: {layers: 3} -# # note units is the head hidden_dim -# units: 256 -# reward_head: {layers: 3} -# cont_head: {layers: 3} -# imag_gradient: 'reinforce' -# batch_size: 256 -# batch_length: 32 -# time_limit: 4000 +craftax_small: + task: craftax_Craftax-Symbolic-AutoReset-v1 + step: 1e6 + action_repeat: 1 + envs: 1 + train_ratio: 512 + video_pred_log: true + dyn_hidden: 256 + dyn_deter: 512 + dyn_stoch: 24 + dyn_discrete: 24 + # note: depth is cnn hidden_dim + encoder: {cnn_keys: 'state_map', cnn_depth: 48, kernel_size: 4, minres: 2, mlp_keys: "state_inventory", mlp_layers: 2, mlp_units: 128} + decoder: {cnn_keys: 'state_map', cnn_depth: 48, kernel_size: 4, minres: 2, mlp_keys: "state_inventory", mlp_layers: 2, mlp_units: 128} + actor: {layers: 3, dist: 'onehot', std: 'none'} + value: {layers: 3} + # note units is the head hidden_dim + units: 256 + reward_head: {layers: 4} + cont_head: {layers: 4} + imag_gradient: 'reinforce' + batch_size: 96 + batch_length: 48 + time_limit: 6000 + dataset_size: 20_000 + precision: 16 + expl_behavior: plan2explore craftax_smaller: task: craftax_Craftax-Symbolic-AutoReset-v1 diff --git a/justfile b/justfile index 125e3de..d563ccd 100644 --- a/justfile +++ b/justfile @@ -4,9 +4,13 @@ set export export OSTYPE := "linux-gnu" export TQDM_MININTERVAL := "30" -main: +craftax_smaller: . ./.venv/bin/activate python dreamer.py --configs craftax_smaller --logdir ./logdir/craftax_smaller + +craftax_small: + . ./.venv/bin/activate + python dreamer.py --configs craftax_small --logdir ./logdir/craftax_small2 logs: tensorboard --logdir logdir/craftax