Skip to content

Actions: Motphys/UniLab

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
ci.yml
1,448 workflow runs
1,448 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
fix(tests): stabilize CUDA MPS CLI tests on macOS CI #1414: Pull request #2511 opened by TATP-233
15m 43s fix/issue-2510-macos-mps-cli-tests
fix/issue-2510-macos-mps-cli-tests
15m 43s
fix(rl): initialize off-policy MPS inference timer CI #1413: Pull request #2509 opened by TATP-233
8m 12s fix/issue-2508-mps-offpolicy-timer
fix/issue-2508-mps-offpolicy-timer
8m 12s
fix(rl): defer off-policy live terminal display until startup completes CI #1412: Pull request #2507 opened by TATP-233
16m 32s fix/issue-2506-terminal-logger-startup-timing
fix/issue-2506-terminal-logger-startup-timing
16m 32s
fix(training): normalize CUDA visibility to UUIDs for MPS launches CI #1411: Pull request #2505 opened by TATP-233
15m 49s fix/issue-2504-mps-uuid-visibility
fix/issue-2504-mps-uuid-visibility
15m 49s
fix(training): handle Ctrl+C cooperatively CI #1410: Pull request #2503 opened by TATP-233
9m 23s fix/issue-2502-graceful-training-interrupt
fix/issue-2502-graceful-training-interrupt
9m 23s
build(deps): prepare wheel-variant migration (#2500) CI #1409: Pull request #2501 opened by TATP-233
4m 33s feat/issue-2500-wheel-variants
feat/issue-2500-wheel-variants
4m 33s
perf(flashsac): add DP graphs and optional BF16 gradient communication CI #1408: Pull request #2499 synchronize by Cudrix-dev
6m 54s Cudrix-dev:perf/flashsac-dp-mps
Cudrix-dev:perf/flashsac-dp-mps
6m 54s
perf(flashsac): add DP graphs and optional BF16 gradient communication CI #1407: Pull request #2499 synchronize by Cudrix-dev
5m 57s Cudrix-dev:perf/flashsac-dp-mps
Cudrix-dev:perf/flashsac-dp-mps
5m 57s
perf(flashsac): add DP graphs and optional BF16 gradient communication CI #1406: Pull request #2499 ready_for_review by Cudrix-dev
7m 42s Cudrix-dev:perf/flashsac-dp-mps
Cudrix-dev:perf/flashsac-dp-mps
7m 42s
perf(flashsac): add DP graphs and optional BF16 gradient communication CI #1405: Pull request #2499 opened by Cudrix-dev
5m 53s Cudrix-dev:perf/flashsac-dp-mps
Cudrix-dev:perf/flashsac-dp-mps
5m 53s
CI CI #1404: Manually run by TATP-233
4m 36s main
main
4m 36s
release: unilab 1.5.0 CI #1403: Pull request #2498 opened by TATP-233
6m 51s release/1.5.0
release/1.5.0
6m 51s
fix(envs): restore mjwarp collector Backend Step timing and surface CUDA MPS status CI #1402: Pull request #2497 synchronize by TATP-233
4m 53s fix/issue-2495-mjwarp-backend-step-timing
fix/issue-2495-mjwarp-backend-step-timing
4m 53s
fix(envs): restore mjwarp collector Backend Step timing and surface CUDA MPS status CI #1401: Pull request #2497 opened by TATP-233
3m 18s fix/issue-2495-mjwarp-backend-step-timing
fix/issue-2495-mjwarp-backend-step-timing
3m 18s
chore(deps): 消费 unisim-core 2.1.0，interval op payload 对齐 torch-first 契约 CI #1400: Pull request #2496 opened by TATP-233
6m 6s chore/unisim-core-2.1.0
chore/unisim-core-2.1.0
6m 6s
refactor(reset)!: remove the v1.4 bool-mask reset compatibility layer (#2474) CI #1399: Pull request #2494 opened by TATP-233
6m 8s fix/issue-2474-reset-env-ids-compat
fix/issue-2474-reset-env-ids-compat
6m 8s
[2474-P2] 清理 env/command 静默缝：删孤儿 hook、metric fail-closed、注释改写 CI #1398: Pull request #2493 opened by TATP-233
12m 44s fix/issue-2474-p2-env-command-silent-seams
fix/issue-2474-p2-env-command-silent-seams
12m 44s
chore(benchmark): delete device_info re-export shim, import unilab.utils.device directly (#2474) CI #1397: Pull request #2491 opened by TATP-233
16m 46s fix/issue-2474-benchmark-device-info-shim
fix/issue-2474-benchmark-device-info-shim
16m 46s
refactor(utils): default reward target_field to 'rewards' and drop 'reward_config' alias (#2474) CI #1396: Pull request #2489 opened by TATP-233
17m 49s fix/issue-2474-reward-target-alias
fix/issue-2474-reward-target-alias
17m 49s
perf(env): collapse duplicate action/control scalar syncs on collector hot path CI #1395: Pull request #2486 opened by TATP-233
8m 16s perf/issue-2482-action-scalar-reads
perf/issue-2482-action-scalar-reads
8m 16s
feat(training): forbid CPU carrier with CUDA learner behind explicit opt-in (#2483) CI #1394: Pull request #2485 opened by TATP-233
4m 22s feat/issue-2483-cpu-carrier-tightening
feat/issue-2483-cpu-carrier-tightening
4m 22s
refactor(training): merge placement knobs into training.collector_tensor_device (#2481) CI #1393: Pull request #2484 opened by TATP-233
5m 51s feat/issue-2481-collector-tensor-device
feat/issue-2481-collector-tensor-device
5m 51s
feat(training): record ambient CUDA MPS state in off-policy probe (#2479) CI #1392: Pull request #2480 synchronize by TATP-233
7m 52s feat/issue-2479-mps-observability
feat/issue-2479-mps-observability
7m 52s
feat(training): record ambient CUDA MPS state in off-policy probe (#2479) CI #1391: Pull request #2480 opened by TATP-233
8m 28s feat/issue-2479-mps-observability
feat/issue-2479-mps-observability
8m 28s
feat(training): add opt-in TF32 matmul knob for PPO/APPO FP32 updates CI #1390: Pull request #2478 opened by TATP-233
9m 22s feat/issue-2475-ppo-appo-tf32
feat/issue-2475-ppo-appo-tf32
9m 22s