refactor!: remove sharpa task and hora algorithm (migrate to sharpa_rl_unilab)

docs/sphinx/source/api_reference/algos/index.md

@@ -8,7 +8,6 @@ published on PyPI; issue #1480):
 - `uni_rl.algos.appo` — APPO runner, learner, staging, worker
 - `uni_rl.algos.fast_sac` / `uni_rl.algos.fast_td3` / `uni_rl.algos.flash_sac` — off-policy learners and runners
 - `uni_rl.offpolicy` — generic off-policy runner, worker, thread budget
-- `uni_rl.algos.hora` — HORA models, trainers, and distillation
 - `uni_rl.algos.common` — shared actor factory, networks, normalization, compile helpers
 
 UniLab keeps the training *entrypoints* (`src/unilab/scripts/train_*.py`),

docs/sphinx/source/api_reference/tasks/index.md

@@ -3,7 +3,7 @@
 Concrete RL tasks split by family:
 
 - **locomotion** — A2, Go1, Go2, Go2w, Go2 + Airbot, and Unitree G1
-- **manipulation** — Allegro / Sharpa in-hand cube and Stewart balance
+- **manipulation** — Allegro in-hand cube and Stewart balance
 - **motion_tracking** — G1 and X2 whole-body motion tracking
 
 Every task is registered into the task `Registry` so it can be selected via

docs/sphinx/source/api_reference/tasks/manipulation.md

@@ -7,6 +7,5 @@
    :recursive:
 
    unilab.tasks.manipulation.allegro_inhand
-   unilab.tasks.manipulation.sharpa_inhand
    unilab.tasks.manipulation.stewart
 ```

docs/sphinx/source/en/0-index.md

@@ -108,7 +108,7 @@ capability requirements.
 :::{grid-item-card} Pick an algorithm
 :link: 2-user_guide/2-algorithms/0-index
 :link-type: doc
-Compare PPO, APPO, SAC, TD3, FlashSAC, and HORA entrypoints.
+Compare PPO, APPO, SAC, TD3, and FlashSAC entrypoints.
 :::
 
 :::{grid-item-card} Deploy or switch sims
@@ -158,7 +158,6 @@ committed benchmark manifest or separate recommendation metadata.
 | Go2W joystick | PPO | PPO owner YAMLs exist for MuJoCo and Motrix flat/rough variants under `src/unilab/conf/ppo/task/go2w_joystick_*`. |
 | G1 locomotion / tracking | PPO, APPO, SAC, TD3 | PPO, APPO, and SAC include committed MuJoCo and Motrix owner YAMLs for G1 tasks; TD3 has a `g1_walk_flat` MuJoCo owner. |
 | Allegro in-hand | PPO, APPO | PPO and APPO have committed MuJoCo and Motrix owner YAMLs for Allegro in-hand tasks. |
-| Sharpa in-hand | PPO, APPO HORA teacher, HORA distillation | Sharpa owner YAMLs are committed for PPO/APPO teacher paths; student distillation uses `src/unilab/conf/hora_distill/task/sharpa_inhand/mujoco.yaml`. |
 
 ```{toctree}
 :hidden:

docs/sphinx/source/en/1-getting_started/2-installation.md

@@ -134,9 +134,8 @@ ROCm/XPU profiles, prefer the uv paths above. Robot meshes and
 textures are intentionally excluded from the wheel and downloaded on the cold
 path from the `unilabsim/unilab-robots` dataset. Ensure the installed package
 location is writable, or pre-fetch assets with `uv run unilab-pull-assets` from a
-source checkout. The isaacgym / isaacsim backends and the HORA multi-GPU
-submission path still assume a source checkout; use their dedicated setup pages
-below.
+source checkout. The isaacgym / isaacsim backends still assume a source
+checkout; use their dedicated setup pages below.
 
 ## Runtime Assets
 

docs/sphinx/source/en/1-getting_started/3-evaluation_and_playback.md

@@ -39,7 +39,7 @@ Use `uv run eval` for regular evaluation and video export. When you need a live
 `src/unilab/scripts/play_interactive.py` script.
 
 `src/unilab/scripts/play_interactive.py` is the general MuJoCo viewer entrypoint for PPO,
-APPO, SAC, FlashSAC, and HORA distill policies. It uses `--algo / --task /
+APPO, SAC, and FlashSAC policies. It uses `--algo / --task /
 --sim` to select the algorithm and owner config. The viewer is always
 `mujoco.viewer`; `--sim` only selects which config to read.
 

docs/sphinx/source/en/1-getting_started/4-project_structure.md

@@ -10,7 +10,7 @@ changing behavior.
 | `src/unilab/conf/` | Hydra roots and task owner YAMLs. The top-level CLI exposes backend selection as `--task` plus `--sim`, then composes the matching owner YAML. |
 | `src/unilab/base/` | Registry, env state, scene, and backend contracts. |
 | `src/unilab/envs/` | Task env implementations and task-specific reset, reward, observation, and DR logic. |
-| `uni_rl` (unilab-rl repo) | PPO, APPO, off-policy, and HORA algorithm code. |
+| `uni_rl` (unilab-rl repo) | PPO, APPO, and off-policy algorithm code. |
 | `uni_rl.ipc` (unilab-rl repo) | Shared-memory and async runner primitives. |
 | `src/unilab/training/` | Shared training helpers for logging, playback, seed handling, and config guards. |
 | `src/unilab/visualization/` | Playback, rendering, NaN inspection, and scene/export utilities. |
@@ -27,7 +27,6 @@ The main config roots are:
 - `src/unilab/conf/sac/config.yaml`, `src/unilab/conf/td3/config.yaml`, and
   `src/unilab/conf/flashsac/config.yaml` for SAC, TD3, and FlashSAC, each with its
   algorithm hyperparameters inlined.
-- `src/unilab/conf/hora_distill/config.yaml` for HORA student distillation.
 
 Task owner YAMLs are the backend identity. Examples:
 

docs/sphinx/source/en/2-user_guide/0-index.md

@@ -15,7 +15,7 @@ CLI routes, Hydra owner YAMLs, logs, checkpoints, and Docker.
 :::{grid-item-card} Algorithms
 :link: 2-algorithms/0-index
 :link-type: doc
-Compare PPO, APPO, SAC, TD3, FlashSAC, and HORA.
+Compare PPO, APPO, SAC, TD3, and FlashSAC.
 :::
 
 :::{grid-item-card} Backends
@@ -42,12 +42,6 @@ Configure reset, init, and interval randomization through task owner configs.
 Export ONNX, inspect NaNs, send W&B logs, and export scenes.
 :::
 
-:::{grid-item-card} Manipulation Notes
-:link: 8-manipulation/0-index
-:link-type: doc
-Task-specific Allegro, Sharpa, and Go2+Airbot notes.
-:::
-
 ::::
 
 ```{toctree}
@@ -61,5 +55,4 @@ Task-specific Allegro, Sharpa, and Go2+Airbot notes.
 5-domain_randomization/0-index
 6-terrain/0-index
 7-tooling/0-index
-8-manipulation/0-index
 ```

docs/sphinx/source/en/2-user_guide/1-training/2-hydra_config.md

@@ -10,7 +10,6 @@ identity of the task, backend, reward, scene, and task-specific runtime fields.
 | PPO | `src/unilab/conf/ppo/task/<task>/<backend>.yaml` |
 | APPO | `src/unilab/conf/appo/task/<task>/<backend>.yaml` |
 | SAC / TD3 / FlashSAC | `src/unilab/conf/<algo>/task/<task>/<backend>.yaml` |
-| HORA distillation | `src/unilab/conf/hora_distill/task/<task>/<backend>.yaml` |
 
 Examples:
 

docs/sphinx/source/en/2-user_guide/1-training/3-logging.md

@@ -74,7 +74,7 @@ Normally the rows sum to `Iter Wall`, so their displayed percentages sum to abou
 hides applicable phases below a 1% threshold: a zero row is kept so it can be matched
 directly with TensorBoard / W&B. Algorithm-specific phases neither occupy terminal
 rows nor get persisted for other algorithms; for example, `Replay Stage` and
-`Weight Publish` exist only for APPO / HORA-APPO.
+`Weight Publish` exist only for APPO.
 
 ### Learner Main Timeline
 

docs/sphinx/source/en/2-user_guide/2-algorithms/0-index.md

@@ -11,7 +11,6 @@ lives, and which command shape selects it. For general flags, see
 | SAC | off-policy | `src/unilab/scripts/train_sac.py` | `src/unilab/conf/sac/config.yaml` |
 | TD3 | off-policy | `src/unilab/scripts/train_td3.py` | `src/unilab/conf/td3/config.yaml` |
 | FlashSAC | off-policy | `src/unilab/scripts/train_flashsac.py` | `src/unilab/conf/flashsac/config.yaml` |
-| HORA | teacher/student distillation path | `scripts/train_hora_distill.py` | `src/unilab/conf/hora_distill/config.yaml` |
 
 ```{toctree}
 :hidden:
@@ -21,5 +20,4 @@ lives, and which command shape selects it. For general flags, see
 3-sac
 4-td3
 5-flash_sac
-7-hora
 ```

docs/sphinx/source/en/2-user_guide/4-tasks/0-index.md

@@ -21,7 +21,7 @@ G1 motion tracking, flips, climbs, wall flips, and box tracking.
 :::{grid-item-card} Manipulation
 :link: 3-manipulation
 :link-type: doc
-Allegro and Sharpa in-hand rotation and grasp generation.
+Allegro in-hand rotation and grasp generation.
 :::
 
 ::::

docs/sphinx/source/en/2-user_guide/4-tasks/3-manipulation.md

@@ -5,19 +5,12 @@ Manipulation tasks live in `src/unilab/tasks/manipulation/`.
 ## In-Hand
 
 - `allegro_inhand` and `allegro_inhand_grasp` have MuJoCo and Motrix PPO owners.
-- `sharpa_inhand`, `sharpa_inhand_grasp`, and the `hora` profile for
-  `sharpa_inhand` are MuJoCo owner paths in the current configs.
 
 ```bash
 uv run train --algo ppo --task allegro_inhand --sim mujoco
 uv run train --algo ppo --task allegro_inhand --sim motrix training.no_play=true
-uv run train --algo ppo --task sharpa_inhand --sim mujoco --profile hora training.no_play=true
 ```
 
-HORA student distillation is configured by
-`src/unilab/conf/hora_distill/task/sharpa_inhand/mujoco.yaml`; it is not currently exposed
-as a separate top-level CLI route.
-
 ## Platform Balancing
 
 `stewart_balance` is a 6-DOF parallel (Stewart) platform that balances a free
@@ -34,5 +27,3 @@ training-stable under load.
 ```bash
 uv run train --algo ppo --task stewart_balance --sim motrix training.no_play=true
 ```
-
-See {doc}`../8-manipulation/1-dexterous_inhand` for in-hand task notes.