inspect-ai 0.3.100__py3-none-any.whl → 0.3.102__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- inspect_ai/_cli/eval.py +1 -1
- inspect_ai/_eval/evalset.py +2 -2
- inspect_ai/_view/www/dist/assets/index.css +44 -12
- inspect_ai/_view/www/dist/assets/index.js +1499 -1467
- inspect_ai/_view/www/package.json +4 -4
- inspect_ai/_view/www/src/app/log-view/tabs/grouping.ts +4 -4
- inspect_ai/_view/www/src/app/routing/navigationHooks.ts +22 -25
- inspect_ai/_view/www/src/app/samples/list/SampleList.tsx +17 -5
- inspect_ai/_view/www/src/state/hooks.ts +1 -1
- inspect_ai/_view/www/yarn.lock +21 -27
- inspect_ai/analysis/beta/__init__.py +2 -0
- inspect_ai/dataset/_sources/csv.py +2 -6
- inspect_ai/dataset/_sources/hf.py +2 -6
- inspect_ai/dataset/_sources/json.py +2 -6
- inspect_ai/dataset/_util.py +23 -0
- inspect_ai/model/_openai.py +4 -0
- inspect_ai/model/_openai_responses.py +11 -6
- inspect_ai/model/_openai_web_search.py +9 -2
- inspect_ai/model/_providers/openai.py +11 -3
- inspect_ai/model/_providers/openai_responses.py +5 -1
- inspect_ai/scorer/_reducer/reducer.py +1 -1
- inspect_ai/tool/_tools/_web_search/_google.py +28 -11
- inspect_ai/tool/_tools/_web_search/_tavily.py +11 -1
- {inspect_ai-0.3.100.dist-info → inspect_ai-0.3.102.dist-info}/METADATA +1 -1
- {inspect_ai-0.3.100.dist-info → inspect_ai-0.3.102.dist-info}/RECORD +29 -29
- {inspect_ai-0.3.100.dist-info → inspect_ai-0.3.102.dist-info}/WHEEL +0 -0
- {inspect_ai-0.3.100.dist-info → inspect_ai-0.3.102.dist-info}/entry_points.txt +0 -0
- {inspect_ai-0.3.100.dist-info → inspect_ai-0.3.102.dist-info}/licenses/LICENSE +0 -0
- {inspect_ai-0.3.100.dist-info → inspect_ai-0.3.102.dist-info}/top_level.txt +0 -0
inspect_ai/_cli/eval.py
CHANGED
|
@@ -35,7 +35,7 @@ from .util import (
|
|
|
35
35
|
)
|
|
36
36
|
|
|
37
37
|
MAX_SAMPLES_HELP = "Maximum number of samples to run in parallel (default is running all samples in parallel)"
|
|
38
|
-
MAX_TASKS_HELP = "Maximum number of tasks to run in parallel (default is 1)"
|
|
38
|
+
MAX_TASKS_HELP = "Maximum number of tasks to run in parallel (default is 1 for eval and 4 for eval-set)"
|
|
39
39
|
MAX_SUBPROCESSES_HELP = (
|
|
40
40
|
"Maximum number of subprocesses to run in parallel (default is os.cpu_count())"
|
|
41
41
|
)
|
inspect_ai/_eval/evalset.py
CHANGED
|
@@ -168,7 +168,7 @@ def eval_set(
|
|
|
168
168
|
max_samples: Maximum number of samples to run in parallel
|
|
169
169
|
(default is max_connections)
|
|
170
170
|
max_tasks: Maximum number of tasks to run in parallel
|
|
171
|
-
(defaults to number of models being evaluated)
|
|
171
|
+
(defaults to the greater of 4 and the number of models being evaluated)
|
|
172
172
|
max_subprocesses: Maximum number of subprocesses to
|
|
173
173
|
run in parallel (default is os.cpu_count())
|
|
174
174
|
max_sandboxes: Maximum number of sandboxes (per-provider)
|
|
@@ -277,7 +277,7 @@ def eval_set(
|
|
|
277
277
|
retry_connections = retry_connections or 0.5
|
|
278
278
|
retry_cleanup = retry_cleanup is not False
|
|
279
279
|
max_connections = starting_max_connections(models, GenerateConfig(**kwargs))
|
|
280
|
-
max_tasks = max_tasks if max_tasks is not None else len(models)
|
|
280
|
+
max_tasks = max_tasks if max_tasks is not None else max(len(models), 4)
|
|
281
281
|
|
|
282
282
|
# prepare console/status
|
|
283
283
|
console = rich.get_console()
|