RoboCasa last-two policies + tokenizers
Datasets + Wave1 logs: hackhackhack66666/rc-last-two
Guides (branch robocasa):
Inventory
| File | Role |
|---|---|
policies/robocasa_turn_off_sink_faucet_topk_ep0500_sr0.580.ckpt |
locked TopK policy (sink) |
policies/robocasa_turn_off_sink_faucet_topk_lock.txt |
lock metadata |
policies/robocasa_turn_off_microwave_topk_ep0500_sr0.620.ckpt |
locked TopK policy (microwave) |
policies/robocasa_turn_off_microwave_topk_lock.txt |
lock metadata |
tokenizers/sink_ep-3080_mse-0.002.ckpt |
frozen OAT tokenizer |
tokenizers/microwave_ep-2720_mse-0.002.ckpt |
frozen OAT tokenizer |
Mentee: enough for AWR16?
Yes, together with rc-last-two (zarr + full Wave1 baselines for both tasks):
- Copy policies โ
my_models/ - Place tokenizers at the paths in the guides (
output/20260720/.../checkpoints/) - Untar zarrs โ
data/robocasa/ - Rsync
wave1_*โoutput/eval/matched_s10000/robocasa/<task>/ - Follow guide: BoN16/32 eval โ collect
--bon_n 16โ train--epochs 100
Wave1 BoN8 for microwave may still be running โ not required to start AWR16.