Press By Number

RoboDojo simulation reference

SE3 Labs real setup

Task ID
SE3 language instruction	“Press the count button times, press the right dark-blue confirm button, press the count button times, press the right dark-blue confirm button again.”
Official language instruction	Press the two red buttons the required number of times according to the number cards, then press the blue button to confirm.
RoboDojo category	Memory
Materials	wooden number cards showing and , each above its count button; three large push buttons colored green/cyan/dark blue from left to right.
Difference	the count buttons are not red. The official high-level instruction mentions one final confirmation, while the official scoring rule requires confirmation after each count; use the scoring rule because it is more specific.
RoboDojo task page	Press By Number

Progress score

Score the highest milestone whose condition holds at the end of the episode; the episode succeeds at 100. Also record the failure mode and the execution metrics below.

Progress	Condition
0	No milestone below is reached.
10	The first press is on the count button.
30	The count button was pressed exactly times with no other press.
45	Then the right confirm button was pressed once.
75	Then the count button was pressed exactly times with no other press.
100	Then the right confirm button was pressed once, with no extra, missing, or out-of-order press. Both arms are back at origin.

Execution metrics

Record these alongside progress; they do not change it. Count over the whole episode, including events later corrected.

Metric	What to count	Answer
First count presses	Registered presses of the count button; target .	exact count
Second count presses	Registered presses of the count button; target .	exact count
Confirm presses	Registered presses of the right confirm button; target 2.	exact count
Order kept	Did every press follow → confirm → → confirm?	yes · no

Randomization

Each episode samples every variable below and records its values.

Variable	Domain
,	The left and middle count buttons, in either order; the right button always confirms.
,	Two different numbers from 1–7 or 9, each shown on the card above its count button.

Python SDK

from se3labs.sdk.eval.tasks import TaskRegistry

definition = TaskRegistry.get_task("ROBODOJO-REAL-17")
# Sample one episode's parameters and initial layout.
task = definition.sample(seed=42)
# Sample 20 episodes with separate seeds; seed=42 reproduces the same set.
tasks = definition.sample_set(seed=42, episodes=20)

Create and run a job:

from se3labs.sdk import eval_client

job = TaskRegistry.get_job(
    "ROBODOJO-REAL-17", policy=policy, seed=42, episodes=20,
)
report = eval_client.serve_policy("your-station", "your-relay:7443", job)

task.layout.visualize("layout.png") saves the sampled layout.