inspect-ai 0.3.80__py3-none-any.whl → 0.3.82__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.
Files changed (179) hide show
  1. inspect_ai/_cli/eval.py +35 -2
  2. inspect_ai/_cli/util.py +44 -1
  3. inspect_ai/_display/core/config.py +1 -1
  4. inspect_ai/_display/core/display.py +13 -4
  5. inspect_ai/_display/core/results.py +1 -1
  6. inspect_ai/_display/textual/widgets/task_detail.py +5 -4
  7. inspect_ai/_eval/eval.py +38 -1
  8. inspect_ai/_eval/evalset.py +5 -0
  9. inspect_ai/_eval/run.py +5 -2
  10. inspect_ai/_eval/task/log.py +53 -6
  11. inspect_ai/_eval/task/run.py +51 -10
  12. inspect_ai/_util/constants.py +2 -0
  13. inspect_ai/_util/file.py +17 -1
  14. inspect_ai/_util/json.py +36 -1
  15. inspect_ai/_view/server.py +113 -1
  16. inspect_ai/_view/www/App.css +1 -1
  17. inspect_ai/_view/www/dist/assets/index.css +518 -296
  18. inspect_ai/_view/www/dist/assets/index.js +38803 -36307
  19. inspect_ai/_view/www/eslint.config.mjs +1 -1
  20. inspect_ai/_view/www/log-schema.json +13 -0
  21. inspect_ai/_view/www/node_modules/flatted/python/flatted.py +149 -0
  22. inspect_ai/_view/www/package.json +8 -2
  23. inspect_ai/_view/www/src/App.tsx +151 -855
  24. inspect_ai/_view/www/src/api/api-browser.ts +176 -5
  25. inspect_ai/_view/www/src/api/api-vscode.ts +75 -1
  26. inspect_ai/_view/www/src/api/client-api.ts +66 -10
  27. inspect_ai/_view/www/src/api/jsonrpc.ts +2 -0
  28. inspect_ai/_view/www/src/api/types.ts +107 -2
  29. inspect_ai/_view/www/src/appearance/icons.ts +1 -0
  30. inspect_ai/_view/www/src/components/AsciinemaPlayer.tsx +3 -3
  31. inspect_ai/_view/www/src/components/DownloadPanel.tsx +2 -2
  32. inspect_ai/_view/www/src/components/ExpandablePanel.tsx +56 -61
  33. inspect_ai/_view/www/src/components/FindBand.tsx +17 -9
  34. inspect_ai/_view/www/src/components/HumanBaselineView.tsx +1 -1
  35. inspect_ai/_view/www/src/components/JsonPanel.tsx +14 -24
  36. inspect_ai/_view/www/src/components/LargeModal.tsx +2 -35
  37. inspect_ai/_view/www/src/components/LightboxCarousel.tsx +27 -11
  38. inspect_ai/_view/www/src/components/LiveVirtualList.module.css +11 -0
  39. inspect_ai/_view/www/src/components/LiveVirtualList.tsx +177 -0
  40. inspect_ai/_view/www/src/components/MarkdownDiv.tsx +3 -3
  41. inspect_ai/_view/www/src/components/MessageBand.tsx +14 -9
  42. inspect_ai/_view/www/src/components/MorePopOver.tsx +3 -3
  43. inspect_ai/_view/www/src/components/NavPills.tsx +20 -8
  44. inspect_ai/_view/www/src/components/NoContentsPanel.module.css +12 -0
  45. inspect_ai/_view/www/src/components/NoContentsPanel.tsx +20 -0
  46. inspect_ai/_view/www/src/components/ProgressBar.module.css +5 -4
  47. inspect_ai/_view/www/src/components/ProgressBar.tsx +3 -2
  48. inspect_ai/_view/www/src/components/PulsingDots.module.css +81 -0
  49. inspect_ai/_view/www/src/components/PulsingDots.tsx +45 -0
  50. inspect_ai/_view/www/src/components/TabSet.tsx +4 -37
  51. inspect_ai/_view/www/src/components/ToolButton.tsx +3 -4
  52. inspect_ai/_view/www/src/index.tsx +26 -94
  53. inspect_ai/_view/www/src/logfile/remoteLogFile.ts +9 -1
  54. inspect_ai/_view/www/src/logfile/remoteZipFile.ts +30 -4
  55. inspect_ai/_view/www/src/metadata/RenderedContent.tsx +4 -6
  56. inspect_ai/_view/www/src/plan/ScorerDetailView.tsx +1 -1
  57. inspect_ai/_view/www/src/samples/InlineSampleDisplay.module.css +9 -1
  58. inspect_ai/_view/www/src/samples/InlineSampleDisplay.tsx +67 -28
  59. inspect_ai/_view/www/src/samples/SampleDialog.tsx +51 -22
  60. inspect_ai/_view/www/src/samples/SampleDisplay.module.css +4 -0
  61. inspect_ai/_view/www/src/samples/SampleDisplay.tsx +144 -90
  62. inspect_ai/_view/www/src/samples/SampleSummaryView.module.css +4 -0
  63. inspect_ai/_view/www/src/samples/SampleSummaryView.tsx +82 -35
  64. inspect_ai/_view/www/src/samples/SamplesTools.tsx +23 -30
  65. inspect_ai/_view/www/src/samples/chat/ChatMessage.tsx +2 -1
  66. inspect_ai/_view/www/src/samples/chat/ChatMessageRenderer.tsx +1 -1
  67. inspect_ai/_view/www/src/samples/chat/ChatViewVirtualList.tsx +45 -53
  68. inspect_ai/_view/www/src/samples/chat/MessageContent.tsx +4 -1
  69. inspect_ai/_view/www/src/samples/chat/MessageContents.tsx +3 -0
  70. inspect_ai/_view/www/src/samples/chat/messages.ts +34 -0
  71. inspect_ai/_view/www/src/samples/chat/tools/ToolCallView.module.css +3 -0
  72. inspect_ai/_view/www/src/samples/chat/tools/ToolCallView.tsx +10 -1
  73. inspect_ai/_view/www/src/samples/chat/tools/ToolInput.tsx +22 -46
  74. inspect_ai/_view/www/src/samples/descriptor/samplesDescriptor.tsx +25 -17
  75. inspect_ai/_view/www/src/samples/descriptor/score/ObjectScoreDescriptor.tsx +2 -1
  76. inspect_ai/_view/www/src/samples/descriptor/types.ts +6 -5
  77. inspect_ai/_view/www/src/samples/list/SampleFooter.module.css +21 -3
  78. inspect_ai/_view/www/src/samples/list/SampleFooter.tsx +20 -1
  79. inspect_ai/_view/www/src/samples/list/SampleList.tsx +105 -85
  80. inspect_ai/_view/www/src/samples/list/SampleRow.module.css +6 -0
  81. inspect_ai/_view/www/src/samples/list/SampleRow.tsx +27 -14
  82. inspect_ai/_view/www/src/samples/sample-tools/SelectScorer.tsx +29 -18
  83. inspect_ai/_view/www/src/samples/sample-tools/SortFilter.tsx +28 -28
  84. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/SampleFilter.tsx +19 -9
  85. inspect_ai/_view/www/src/samples/sampleDataAdapter.ts +33 -0
  86. inspect_ai/_view/www/src/samples/sampleLimit.ts +2 -2
  87. inspect_ai/_view/www/src/samples/scores/SampleScoreView.tsx +7 -9
  88. inspect_ai/_view/www/src/samples/scores/SampleScores.tsx +7 -11
  89. inspect_ai/_view/www/src/samples/transcript/ErrorEventView.tsx +0 -13
  90. inspect_ai/_view/www/src/samples/transcript/InfoEventView.tsx +0 -13
  91. inspect_ai/_view/www/src/samples/transcript/InputEventView.tsx +0 -13
  92. inspect_ai/_view/www/src/samples/transcript/ModelEventView.module.css +4 -0
  93. inspect_ai/_view/www/src/samples/transcript/ModelEventView.tsx +10 -24
  94. inspect_ai/_view/www/src/samples/transcript/SampleInitEventView.tsx +0 -13
  95. inspect_ai/_view/www/src/samples/transcript/SampleLimitEventView.tsx +4 -22
  96. inspect_ai/_view/www/src/samples/transcript/SandboxEventView.tsx +15 -24
  97. inspect_ai/_view/www/src/samples/transcript/ScoreEventView.tsx +0 -13
  98. inspect_ai/_view/www/src/samples/transcript/StepEventView.tsx +6 -28
  99. inspect_ai/_view/www/src/samples/transcript/SubtaskEventView.tsx +24 -34
  100. inspect_ai/_view/www/src/samples/transcript/ToolEventView.module.css +4 -0
  101. inspect_ai/_view/www/src/samples/transcript/ToolEventView.tsx +8 -13
  102. inspect_ai/_view/www/src/samples/transcript/TranscriptView.tsx +197 -338
  103. inspect_ai/_view/www/src/samples/transcript/TranscriptVirtualListComponent.module.css +16 -0
  104. inspect_ai/_view/www/src/samples/transcript/TranscriptVirtualListComponent.tsx +44 -0
  105. inspect_ai/_view/www/src/samples/transcript/event/EventNav.tsx +7 -4
  106. inspect_ai/_view/www/src/samples/transcript/event/EventPanel.tsx +52 -58
  107. inspect_ai/_view/www/src/samples/transcript/event/EventProgressPanel.module.css +23 -0
  108. inspect_ai/_view/www/src/samples/transcript/event/EventProgressPanel.tsx +27 -0
  109. inspect_ai/_view/www/src/samples/transcript/state/StateEventRenderers.tsx +30 -1
  110. inspect_ai/_view/www/src/samples/transcript/state/StateEventView.tsx +102 -72
  111. inspect_ai/_view/www/src/scoring/utils.ts +87 -0
  112. inspect_ai/_view/www/src/state/appSlice.ts +244 -0
  113. inspect_ai/_view/www/src/state/hooks.ts +397 -0
  114. inspect_ai/_view/www/src/state/logPolling.ts +196 -0
  115. inspect_ai/_view/www/src/state/logSlice.ts +214 -0
  116. inspect_ai/_view/www/src/state/logsPolling.ts +118 -0
  117. inspect_ai/_view/www/src/state/logsSlice.ts +181 -0
  118. inspect_ai/_view/www/src/state/samplePolling.ts +311 -0
  119. inspect_ai/_view/www/src/state/sampleSlice.ts +127 -0
  120. inspect_ai/_view/www/src/state/sampleUtils.ts +21 -0
  121. inspect_ai/_view/www/src/state/scrolling.ts +206 -0
  122. inspect_ai/_view/www/src/state/store.ts +168 -0
  123. inspect_ai/_view/www/src/state/store_filter.ts +84 -0
  124. inspect_ai/_view/www/src/state/utils.ts +23 -0
  125. inspect_ai/_view/www/src/storage/index.ts +26 -0
  126. inspect_ai/_view/www/src/types/log.d.ts +2 -0
  127. inspect_ai/_view/www/src/types.ts +94 -32
  128. inspect_ai/_view/www/src/utils/attachments.ts +58 -23
  129. inspect_ai/_view/www/src/utils/logger.ts +52 -0
  130. inspect_ai/_view/www/src/utils/polling.ts +100 -0
  131. inspect_ai/_view/www/src/utils/react.ts +30 -0
  132. inspect_ai/_view/www/src/utils/vscode.ts +1 -1
  133. inspect_ai/_view/www/src/workspace/WorkSpace.tsx +181 -216
  134. inspect_ai/_view/www/src/workspace/WorkSpaceView.tsx +11 -53
  135. inspect_ai/_view/www/src/workspace/navbar/Navbar.tsx +8 -18
  136. inspect_ai/_view/www/src/workspace/navbar/PrimaryBar.module.css +1 -0
  137. inspect_ai/_view/www/src/workspace/navbar/PrimaryBar.tsx +40 -22
  138. inspect_ai/_view/www/src/workspace/navbar/ResultsPanel.module.css +0 -1
  139. inspect_ai/_view/www/src/workspace/navbar/ResultsPanel.tsx +98 -39
  140. inspect_ai/_view/www/src/workspace/navbar/RunningStatusPanel.module.css +32 -0
  141. inspect_ai/_view/www/src/workspace/navbar/RunningStatusPanel.tsx +32 -0
  142. inspect_ai/_view/www/src/workspace/navbar/SecondaryBar.tsx +11 -13
  143. inspect_ai/_view/www/src/workspace/navbar/StatusPanel.tsx +6 -2
  144. inspect_ai/_view/www/src/workspace/sidebar/LogDirectoryTitleView.tsx +4 -4
  145. inspect_ai/_view/www/src/workspace/sidebar/Sidebar.tsx +28 -13
  146. inspect_ai/_view/www/src/workspace/tabs/InfoTab.tsx +5 -10
  147. inspect_ai/_view/www/src/workspace/tabs/JsonTab.tsx +4 -4
  148. inspect_ai/_view/www/src/workspace/tabs/RunningNoSamples.module.css +22 -0
  149. inspect_ai/_view/www/src/workspace/tabs/RunningNoSamples.tsx +19 -0
  150. inspect_ai/_view/www/src/workspace/tabs/SamplesTab.tsx +110 -115
  151. inspect_ai/_view/www/src/workspace/tabs/grouping.ts +37 -5
  152. inspect_ai/_view/www/src/workspace/tabs/types.ts +4 -0
  153. inspect_ai/_view/www/src/workspace/types.ts +4 -3
  154. inspect_ai/_view/www/src/workspace/utils.ts +4 -4
  155. inspect_ai/_view/www/vite.config.js +6 -0
  156. inspect_ai/_view/www/yarn.lock +370 -354
  157. inspect_ai/log/_condense.py +26 -0
  158. inspect_ai/log/_log.py +6 -3
  159. inspect_ai/log/_recorders/buffer/__init__.py +14 -0
  160. inspect_ai/log/_recorders/buffer/buffer.py +30 -0
  161. inspect_ai/log/_recorders/buffer/database.py +685 -0
  162. inspect_ai/log/_recorders/buffer/filestore.py +259 -0
  163. inspect_ai/log/_recorders/buffer/types.py +84 -0
  164. inspect_ai/log/_recorders/eval.py +2 -11
  165. inspect_ai/log/_recorders/types.py +30 -0
  166. inspect_ai/log/_transcript.py +27 -1
  167. inspect_ai/model/_call_tools.py +1 -0
  168. inspect_ai/model/_generate_config.py +2 -2
  169. inspect_ai/model/_model.py +1 -0
  170. inspect_ai/tool/_tool_support_helpers.py +4 -4
  171. inspect_ai/tool/_tools/_web_browser/_web_browser.py +3 -1
  172. inspect_ai/util/_subtask.py +1 -0
  173. {inspect_ai-0.3.80.dist-info → inspect_ai-0.3.82.dist-info}/METADATA +2 -2
  174. {inspect_ai-0.3.80.dist-info → inspect_ai-0.3.82.dist-info}/RECORD +178 -138
  175. inspect_ai/_view/www/src/samples/transcript/SampleTranscript.tsx +0 -22
  176. {inspect_ai-0.3.80.dist-info → inspect_ai-0.3.82.dist-info}/WHEEL +0 -0
  177. {inspect_ai-0.3.80.dist-info → inspect_ai-0.3.82.dist-info}/entry_points.txt +0 -0
  178. {inspect_ai-0.3.80.dist-info → inspect_ai-0.3.82.dist-info}/licenses/LICENSE +0 -0
  179. {inspect_ai-0.3.80.dist-info → inspect_ai-0.3.82.dist-info}/top_level.txt +0 -0
@@ -1,12 +1,16 @@
1
1
  import { SampleSummary } from "../../api/types";
2
2
  import { SamplesDescriptor } from "../../samples/descriptor/samplesDescriptor";
3
+ import { ScoreLabel } from "../../types";
4
+ import { Epochs } from "../../types/log";
3
5
  import { ListItem, SampleListItem, SeparatorListItem } from "./types";
4
6
 
5
7
  export const getSampleProcessor = (
6
8
  samples: SampleSummary[],
9
+ epochs: Epochs,
7
10
  groupBy: "sample" | "epoch" | "none",
8
11
  groupByOrder: "asc" | "desc",
9
12
  sampleDescriptor: SamplesDescriptor,
13
+ score?: ScoreLabel,
10
14
  ): ((
11
15
  sample: SampleSummary,
12
16
  index: number,
@@ -14,11 +18,17 @@ export const getSampleProcessor = (
14
18
  ) => ListItem[]) => {
15
19
  // Perform grouping if there are epochs
16
20
  if (groupBy == "epoch") {
17
- return groupByEpoch(samples, sampleDescriptor, groupByOrder);
21
+ return groupByEpoch(samples, epochs, sampleDescriptor, groupByOrder, score);
18
22
  } else if (groupBy === "sample") {
19
- return groupBySample(samples, sampleDescriptor, groupByOrder);
23
+ return groupBySample(
24
+ samples,
25
+ epochs,
26
+ sampleDescriptor,
27
+ groupByOrder,
28
+ score,
29
+ );
20
30
  } else {
21
- return noGrouping(samples, groupByOrder);
31
+ return noGrouping(samples, groupByOrder, sampleDescriptor, score);
22
32
  }
23
33
  };
24
34
 
@@ -28,6 +38,8 @@ export const getSampleProcessor = (
28
38
  const noGrouping = (
29
39
  samples: SampleSummary[],
30
40
  order: "asc" | "desc",
41
+ sampleDescriptor: SamplesDescriptor,
42
+ score?: ScoreLabel,
31
43
  ): ((sample: SampleSummary, index: number) => ListItem[]) => {
32
44
  const counter = getCounter(samples.length, 1, order);
33
45
  return (sample: SampleSummary, index: number) => {
@@ -40,6 +52,12 @@ const noGrouping = (
40
52
  index: index,
41
53
  data: sample,
42
54
  type: "sample",
55
+ answer:
56
+ sampleDescriptor.selectedScorerDescriptor(sample)?.answer() || "",
57
+ scoreRendered: sampleDescriptor.evalDescriptor
58
+ .score(sample, score)
59
+ ?.render(),
60
+ completed: sample.completed !== undefined ? sample.completed : true,
43
61
  },
44
62
  ];
45
63
  };
@@ -50,8 +68,10 @@ const noGrouping = (
50
68
  */
51
69
  const groupBySample = (
52
70
  samples: SampleSummary[],
71
+ epochs: Epochs,
53
72
  sampleDescriptor: SamplesDescriptor,
54
73
  order: "asc" | "desc",
74
+ score?: ScoreLabel,
55
75
  ): ((
56
76
  sample: SampleSummary,
57
77
  index: number,
@@ -73,7 +93,7 @@ const groupBySample = (
73
93
  }
74
94
  }
75
95
  });
76
- const groupCount = samples.length / sampleDescriptor.evalDescriptor.epochs;
96
+ const groupCount = samples.length / (epochs || 1);
77
97
  const itemCount = samples.length / groupCount;
78
98
  const counter = getCounter(itemCount, groupCount, order);
79
99
  return (
@@ -103,6 +123,11 @@ const groupBySample = (
103
123
  index: index,
104
124
  data: sample,
105
125
  type: "sample",
126
+ answer: sampleDescriptor.selectedScorerDescriptor(sample)?.answer() || "",
127
+ scoreRendered: sampleDescriptor.evalDescriptor
128
+ .score(sample, score)
129
+ ?.render(),
130
+ completed: sample.completed !== undefined ? sample.completed : true,
106
131
  } as SampleListItem);
107
132
 
108
133
  return results;
@@ -114,14 +139,16 @@ const groupBySample = (
114
139
  */
115
140
  const groupByEpoch = (
116
141
  samples: SampleSummary[],
142
+ epochs: Epochs,
117
143
  sampleDescriptor: SamplesDescriptor,
118
144
  order: "asc" | "desc",
145
+ score?: ScoreLabel,
119
146
  ): ((
120
147
  sample: SampleSummary,
121
148
  index: number,
122
149
  previousSample?: SampleSummary,
123
150
  ) => ListItem[]) => {
124
- const groupCount = sampleDescriptor.evalDescriptor.epochs;
151
+ const groupCount = epochs || 1;
125
152
  const itemCount = samples.length / groupCount;
126
153
  const counter = getCounter(itemCount, groupCount, order);
127
154
 
@@ -153,6 +180,11 @@ const groupByEpoch = (
153
180
  index: index,
154
181
  data: sample,
155
182
  type: "sample",
183
+ answer: sampleDescriptor.selectedScorerDescriptor(sample)?.answer() || "",
184
+ scoreRendered: sampleDescriptor.evalDescriptor
185
+ .score(sample, score)
186
+ ?.render(),
187
+ completed: sample.completed !== undefined ? sample.completed : true,
156
188
  } as SampleListItem);
157
189
 
158
190
  return results;
@@ -1,11 +1,15 @@
1
+ import { ReactNode } from "react";
1
2
  import { SampleSummary } from "../../api/types";
2
3
 
3
4
  export interface SampleListItem {
4
5
  label: string;
5
6
  index: number;
6
7
  number: number;
8
+ answer: string;
9
+ scoreRendered: ReactNode;
7
10
  data: SampleSummary;
8
11
  type: "sample";
12
+ completed: boolean;
9
13
  }
10
14
 
11
15
  export interface SeparatorListItem {
@@ -1,10 +1,11 @@
1
- import { ReactNode, RefObject } from "react";
1
+ import { ComponentType, ReactNode, RefObject } from "react";
2
2
 
3
- export interface TabDescriptor {
3
+ export interface TabDescriptor<P> {
4
4
  id: string;
5
5
  scrollable: boolean;
6
6
  scrollRef?: RefObject<HTMLDivElement | null>;
7
7
  label: string;
8
- content: () => ReactNode;
8
+ component: ComponentType<P>;
9
+ componentProps: P;
9
10
  tools?: () => ReactNode[] | undefined;
10
11
  }
@@ -1,6 +1,6 @@
1
- import { EvalMetric } from "../types/log";
1
+ import { ResultsMetric } from "./navbar/ResultsPanel";
2
2
 
3
- export const metricDisplayName = (metric: EvalMetric): string => {
3
+ export const metricDisplayName = (metric: ResultsMetric): string => {
4
4
  let modifier = undefined;
5
5
  for (const metricModifier of metricModifiers) {
6
6
  modifier = metricModifier(metric);
@@ -13,10 +13,10 @@ export const metricDisplayName = (metric: EvalMetric): string => {
13
13
  return metricName;
14
14
  };
15
15
 
16
- type MetricModifier = (metric: EvalMetric) => string | undefined;
16
+ type MetricModifier = (metric: ResultsMetric) => string | undefined;
17
17
 
18
18
  const clusterMetricModifier: MetricModifier = (
19
- metric: EvalMetric,
19
+ metric: ResultsMetric,
20
20
  ): string | undefined => {
21
21
  if (metric.name !== "stderr") {
22
22
  return undefined;
@@ -24,4 +24,10 @@ export default defineConfig({
24
24
  resolve: {
25
25
  dedupe: ["react", "react-dom"],
26
26
  },
27
+ define: {
28
+ __DEV_WATCH__: JSON.stringify(process.env.DEV_LOGGING === "true"),
29
+ __LOGGING_FILTER__: JSON.stringify(
30
+ process.env.DEV_LOGGING_NAMESPACES || "*",
31
+ ),
32
+ },
27
33
  });