Commit 9fe2ba7
Pins what the failure bound guarantees across parallel environments
Review on #7433 noted that with several environments the recorded failure count can pass
max_num_failures. Measured on the shipped loop: the overshoot is exactly the number of attempts
that end on the step which crosses the bound, minus one, so it is zero for a single environment
and at most num_envs - 1. It cannot be driven to zero -- those attempts have already completed
when the bound is read, and stopping a generator instead leaves env_loop waiting forever for an
action from it, since the loop blocks until all num_envs actions are queued.
So the field's docstring now states the guarantee for parallel runs, and two tests pin both
sides of it: exact when attempts end on separate steps, bounded by num_envs - 1 when they end
together. The test harness gains num_envs and attempts_per_step, defaulting to what it did before.
No behaviour change.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019SsU8ziGULaEBbtmvJSQZW1 parent 3bcd74c commit 9fe2ba7
2 files changed
Lines changed: 42 additions & 10 deletions
File tree
- source
- isaaclab_mimic/test
- isaaclab/isaaclab/envs
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
48 | 48 | | |
49 | 49 | | |
50 | 50 | | |
| 51 | + | |
| 52 | + | |
| 53 | + | |
| 54 | + | |
51 | 55 | | |
52 | 56 | | |
53 | 57 | | |
| |||
Lines changed: 38 additions & 10 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
31 | 31 | | |
32 | 32 | | |
33 | 33 | | |
34 | | - | |
| 34 | + | |
| 35 | + | |
| 36 | + | |
| 37 | + | |
| 38 | + | |
| 39 | + | |
| 40 | + | |
| 41 | + | |
| 42 | + | |
35 | 43 | | |
36 | 44 | | |
| 45 | + | |
| 46 | + | |
| 47 | + | |
37 | 48 | | |
38 | 49 | | |
39 | 50 | | |
| |||
42 | 53 | | |
43 | 54 | | |
44 | 55 | | |
45 | | - | |
| 56 | + | |
| 57 | + | |
46 | 58 | | |
47 | 59 | | |
48 | 60 | | |
49 | 61 | | |
50 | 62 | | |
51 | 63 | | |
52 | 64 | | |
53 | | - | |
54 | | - | |
55 | | - | |
56 | | - | |
57 | | - | |
58 | | - | |
| 65 | + | |
| 66 | + | |
| 67 | + | |
| 68 | + | |
| 69 | + | |
| 70 | + | |
| 71 | + | |
| 72 | + | |
59 | 73 | | |
60 | 74 | | |
61 | | - | |
| 75 | + | |
62 | 76 | | |
63 | | - | |
| 77 | + | |
64 | 78 | | |
65 | 79 | | |
66 | 80 | | |
| |||
120 | 134 | | |
121 | 135 | | |
122 | 136 | | |
| 137 | + | |
| 138 | + | |
| 139 | + | |
| 140 | + | |
| 141 | + | |
| 142 | + | |
| 143 | + | |
| 144 | + | |
| 145 | + | |
| 146 | + | |
| 147 | + | |
| 148 | + | |
| 149 | + | |
| 150 | + | |
0 commit comments