mahgoobi commited on
Commit
8939fb6
·
verified ·
1 Parent(s): f92f62a

add search runs

Browse files
README.md CHANGED
@@ -28,7 +28,7 @@ A branch the search dropped was cut off mid-episode, so it is resumed from its o
28
  and carried to a finish — the action nobody executed still gets an answer to *would this
29
  have worked*.
30
 
31
- 252 searches · 13 tasks · 9,435 nodes, each with its own
32
  state and image.
33
 
34
  This repo hosts the data. **What it means, how it was produced and how to use it live in
@@ -43,7 +43,7 @@ on the other without having seen it.
43
 
44
  | split | seeds | in this repo |
45
  |---|---|---|
46
- | `train` | collection seeds (50000+) — fans to learn from | 252 searches |
47
  | `test` | the benchmark's own evaluation seeds (40000+) — what a scorer is measured on | **not collected yet** |
48
 
49
  ```python
@@ -75,7 +75,7 @@ and should not be pooled. Runs of a task the change did not touch stay comparabl
75
  | `put_can_in_cabinet` | 20 | train | kitchenl_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50020 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
76
  | `put_cup_on_coaster` | 20 | train | study_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50020 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
77
  | `put_milktea_on_shelf` | 20 | train | office_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50019 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
78
- | `put_mouse_on_pad` | 20 | train | office_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50020 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
79
  | `put_phone_on_holder` | 20 | train | office_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50019 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
80
  | `put_stapler_next_to_mouse` | 20 | train | office_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50019 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
81
 
@@ -92,8 +92,8 @@ and should not be pooled. Runs of a task the change did not touch stay comparabl
92
 
93
  | table | rows | files | columns |
94
  |---|---|---|---|
95
- | `nodes` | 9,435 | 252 | 10 |
96
- | `steps` | 459,150 | 252 | 5 |
97
 
98
  Outcomes, best to worst: `hard_success` solved it cleanly, `soft_success` solved it after
99
  a collision, `soft_failure` missed, `hard_failure` missed and collided. `terminal` says
@@ -102,9 +102,9 @@ it is true, and only those are counted here.
102
 
103
  | outcome | branches |
104
  |---|---|
105
- | `hard_success` | 814 |
106
  | `soft_success` | 8 |
107
- | `soft_failure` | 421 |
108
  | `hard_failure` | 17 |
109
 
110
  ## Loading it
 
28
  and carried to a finish — the action nobody executed still gets an answer to *would this
29
  have worked*.
30
 
31
+ 253 searches · 13 tasks · 9,479 nodes, each with its own
32
  state and image.
33
 
34
  This repo hosts the data. **What it means, how it was produced and how to use it live in
 
43
 
44
  | split | seeds | in this repo |
45
  |---|---|---|
46
+ | `train` | collection seeds (50000+) — fans to learn from | 253 searches |
47
  | `test` | the benchmark's own evaluation seeds (40000+) — what a scorer is measured on | **not collected yet** |
48
 
49
  ```python
 
75
  | `put_can_in_cabinet` | 20 | train | kitchenl_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50020 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
76
  | `put_cup_on_coaster` | 20 | train | study_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50020 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
77
  | `put_milktea_on_shelf` | 20 | train | office_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50019 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
78
+ | `put_mouse_on_pad` | 21 | train | office_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50019, 50020 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
79
  | `put_phone_on_holder` | 20 | train | office_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50019 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
80
  | `put_stapler_next_to_mouse` | 20 | train | office_clean | 50000, 50001, 50002, 50003, 50004, 50005, 50006, 50007, 50008, 50009, 50010, 50011, 50012, 50013, 50014, 50015, 50016, 50017, 50018, 50019 | [RoboPRO @ `2a1adee`](https://github.com/EAI-RSM/RoboPRO/commit/2a1adee49dbaebe06379398745233b8f0b43c85f) |
81
 
 
92
 
93
  | table | rows | files | columns |
94
  |---|---|---|---|
95
+ | `nodes` | 9,479 | 253 | 10 |
96
+ | `steps` | 461,300 | 253 | 5 |
97
 
98
  Outcomes, best to worst: `hard_success` solved it cleanly, `soft_success` solved it after
99
  a collision, `soft_failure` missed, `hard_failure` missed and collided. `terminal` says
 
102
 
103
  | outcome | branches |
104
  |---|---|
105
+ | `hard_success` | 816 |
106
  | `soft_success` | 8 |
107
+ | `soft_failure` | 424 |
108
  | `hard_failure` | 17 |
109
 
110
  ## Loading it
data/nodes/task=put_mouse_on_pad/put_mouse_on_pad__bench_demo_office_clean__seed50019.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04bffd54a0ef7bf0fe6a019c5809003531fce6a45ebc9df49e59a9e74025ba34
3
+ size 1544403
data/steps/task=put_mouse_on_pad/put_mouse_on_pad__bench_demo_office_clean__seed50019.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d1fde31d5da46cd0453bad1d5f0e2a71781ef3759b9b6d586415a941b452da2
3
+ size 770012
meta/configs/put_mouse_on_pad__bench_demo_office_clean__seed50019.yml ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ scene:
2
+ task: put_mouse_on_pad
3
+ task_config: bench_demo_office_clean
4
+ seed: 50019
5
+ seeds: null
6
+ eval_mode: true
7
+ policy:
8
+ name: roboresearch_policy
9
+ config: /work/mohammed/roboresearch/critic_fans/deploy_fan5.yml
10
+ instruction: null
11
+ instruction_type: unseen
12
+ candidate_opt:
13
+ seed: 338752743
14
+ overrides: {}
15
+ search:
16
+ strategy: branch_once
17
+ k: 5
18
+ depth: -1
19
+ top_m: 1
20
+ stop_on_success: false
21
+ params: {}
22
+ fitness:
23
+ name: oracle
24
+ dist_tiebreak: true
25
+ verify:
26
+ enabled: false
27
+ atol: 1.0e-05
28
+ output:
29
+ plans: /work/mohammed/roboresearch/critic_fans/put_mouse_on_pad/plans
30
+ episodes: null
31
+ record:
32
+ dir: /work/mohammed/roboresearch/critic_fans/record
33
+ save_intermediate: false
34
+ cameras: []
35
+ carry_branches: true
36
+ branch_every: 0
37
+ carry_fixed_noise: false
meta/searches/put_mouse_on_pad__bench_demo_office_clean__seed50019.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0e58e713b83ddd3db7b17a1d5c73a34c847af7f4a69b042d47940e8a8988e34
3
+ size 14370