Shiki42/s016-sortblocks-concurrent-act-step100000

S016 Blocks Ranking Concurrent ACT, step 100000

Official LeRobot 0.4.4 ACT policy trained for the RoboTwin Blocks Ranking RGB-CTR simulation task. This repository contains the preselected final inference checkpoint from CTR Experiment E768, Run E768-R001.

Training identity

• Training dataset: Shiki42/ctr-sortblocks-100ep-concurrent at immutable revision b93f1c7b54f2b17110349049ad8f0a091493acd0 (100 episodes, 45,959 frames).
• ACT training: 100,000 optimizer updates, effective batch size 8, seed 87431, Sequential/Concurrent study S016's Concurrent arm.
• Source: official LeRobot 0.4.4 commit 8fff0fde7c79f23a93d845d1a50e985de01f8b8a; CTR launch code commit 418d5115033ecaea8c8b88a509b2ce6ee6.
• Non-CTR dataset: IdleMask disabled. Saved LeRobot preprocessor and postprocessor provide the training normalization and inverse transform.
• Cameras: top and centered fovy90 left/right wrist RGB; state and action vectors have 14 dimensions.

Checkpoint and status

The formal Run completed with exit code 0 on 2026-09-23 after all 100,000 updates. The final checkpoint passed a fresh-process CPU reload: 234 finite model state tensors matched model.safetensors exactly, and both saved processors loaded. The model file SHA-256 is af879c951151963abc0f2da998921c8a7511000eca1ca7b978402cd888672541. The complete public inference package also includes SHA256SUMS.

Held-out RoboTwin success and collision rates have not been established. The shared held-out 100-scene suite and ACT evaluation adapter are still awaiting qualification (evaluation Experiment E783). Do not treat training completion as a task-success result.

Optimizer and RNG training state are retained in the external formal Run artifacts and are not part of this inference repository.