Built site for gh-pages

UKGovernmentBEIS · Mar 6, 2025 · 8d32391 · 8d32391
1 parent 7709598
commit 8d32391
Show file tree

Hide file tree

Showing 64 changed files with 4,037 additions and 2,192 deletions.
diff --git a/.nojekyll b/.nojekyll
@@ -1 +1 @@
-a322beae
+9bf1da78
diff --git a/agent-bridge.html b/agent-bridge.html
@@ -301,6 +301,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">

diff --git a/agents-api.html b/agents-api.html
@@ -301,6 +301,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">

diff --git a/agents.html b/agents.html
@@ -301,6 +301,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">

diff --git a/approval.html b/approval.html
@@ -301,6 +301,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">

diff --git a/caching.html b/caching.html
@@ -301,6 +301,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">

diff --git a/datasets.html b/datasets.html
@@ -301,6 +301,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">

diff --git a/errors-and-limits.html b/errors-and-limits.html
@@ -301,6 +301,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">

diff --git a/eval-logs.html b/eval-logs.html
@@ -304,6 +304,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">
@@ -1385,7 +1388,7 @@ <h3 class="anchored" data-anchor-id="log-schema">Log Schema</h3>
     </div>
   </div>
 </footer>
-<script>var lightboxQuarto = GLightbox({"openEffect":"zoom","descPosition":"bottom","selector":".lightbox","closeEffect":"zoom","loop":false});
+<script>var lightboxQuarto = GLightbox({"openEffect":"zoom","loop":false,"selector":".lightbox","descPosition":"bottom","closeEffect":"zoom"});
 window.onload = () => {
   lightboxQuarto.on('slide_before_load', (data) => {
     const { slideIndex, slideNode, slideConfig, player, trigger } = data;

diff --git a/eval-sets.html b/eval-sets.html
@@ -301,6 +301,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">

diff --git a/extensions.html b/extensions.html
@@ -300,6 +300,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">
@@ -648,40 +651,45 @@ <h2 class="anchored" data-anchor-id="sec-sandbox-environment-extensions">Sandbox
 </thead>
 <tbody>
 <tr class="odd">
-<td><code>config_files()</code></td>
-<td>Called once to determine the names of ‘default’ config files for this provider (e.g.&nbsp;‘compose.yaml’).</td>
-<td></td>
-</tr>
-<tr class="even">
 <td><code>default_concurrency()</code></td>
-<td>Called once to determine the default maximum number of sandboxes to run in parallel. Return <code>None</code> for no limit (the default behavior).</td>
+<td>Called once to determine the default maximum number of sandboxes to run in parallel. Return <code>None</code> for no limit (the default behaviour).</td>
 <td></td>
 </tr>
-<tr class="odd">
+<tr class="even">
 <td><code>task_init()</code></td>
 <td>Called once for each unique sandbox environment config before executing the tasks in an <span class="element-type-name ref-interlink"><a href="./reference/inspect_ai.html#eval">eval()</a></span> run.</td>
 <td>Expensive initialisation operations (e.g.&nbsp;pulling or building images)</td>
 </tr>
-<tr class="even">
+<tr class="odd">
 <td><code>sample_init()</code></td>
 <td>Called at the beginning of each <span class="element-type-name ref-interlink"><a href="./reference/inspect_ai.dataset.html#sample">Sample</a></span>.</td>
 <td>Create <span class="element-type-name ref-interlink"><a href="./reference/inspect_ai.util.html#sandboxenvironment">SandboxEnvironment</a></span> instances for the sample.</td>
 </tr>
-<tr class="odd">
+<tr class="even">
 <td><code>sample_cleanup()</code></td>
 <td>Called at the end of each <span class="element-type-name ref-interlink"><a href="./reference/inspect_ai.dataset.html#sample">Sample</a></span></td>
 <td>Cleanup <span class="element-type-name ref-interlink"><a href="./reference/inspect_ai.util.html#sandboxenvironment">SandboxEnvironment</a></span> instances for the sample.</td>
 </tr>
-<tr class="even">
+<tr class="odd">
 <td><code>task_cleanup()</code></td>
 <td>Called once for each unique sandbox environment config after executing the tasks in an <span class="element-type-name ref-interlink"><a href="./reference/inspect_ai.html#eval">eval()</a></span> run.</td>
 <td>Last chance handler for any resources not yet cleaned up (see also discussion below).</td>
 </tr>
-<tr class="odd">
+<tr class="even">
 <td><code>cli_cleanup()</code></td>
 <td>Called via <code>inspect sandbox cleanup</code></td>
 <td>CLI invoked manual cleanup of resources created by this <span class="element-type-name ref-interlink"><a href="./reference/inspect_ai.util.html#sandboxenvironment">SandboxEnvironment</a></span>.</td>
 </tr>
+<tr class="odd">
+<td><code>config_files()</code></td>
+<td>Called once to determine the names of ‘default’ config files for this provider (e.g.&nbsp;‘compose.yaml’).</td>
+<td></td>
+</tr>
+<tr class="even">
+<td><code>config_deserialize()</code></td>
+<td>Called when a custom sandbox config type is read from a log file.</td>
+<td>Only required if a sandbox supports custom config types.</td>
+</tr>
 </tbody>
 </table>
 <p>In the case of parallel execution of a group of tasks within the same working directory, the <code>task_init()</code> and <code>task_cleanup()</code> functions will be called once for each unique sandbox environment configuration (e.g.&nbsp;Docker Compose file). This is a performance optimisation derived from the fact that initialisation and cleanup are shared for tasks with identical configurations.</p>

diff --git a/extensions.html.md b/extensions.html.md
@@ -240,15 +240,16 @@ registration of sandboxes from the importing of libraries they require
 The class methods take care of various stages of initialisation, setup,
 and teardown:
 
-| Method                  | Lifecycle                                                                                                                               | Purpose                                                                               |
-|-------------------------|-----------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------|
-| `config_files()`        | Called once to determine the names of ‘default’ config files for this provider (e.g. ‘compose.yaml’).                                   |                                                                                       |
-| `default_concurrency()` | Called once to determine the default maximum number of sandboxes to run in parallel. Return `None` for no limit (the default behavior). |                                                                                       |
-| `task_init()`           | Called once for each unique sandbox environment config before executing the tasks in an `eval()` run.                                   | Expensive initialisation operations (e.g. pulling or building images)                 |
-| `sample_init()`         | Called at the beginning of each `Sample`.                                                                                               | Create `SandboxEnvironment` instances for the sample.                                 |
-| `sample_cleanup()`      | Called at the end of each `Sample`                                                                                                      | Cleanup `SandboxEnvironment` instances for the sample.                                |
-| `task_cleanup()`        | Called once for each unique sandbox environment config after executing the tasks in an `eval()` run.                                    | Last chance handler for any resources not yet cleaned up (see also discussion below). |
-| `cli_cleanup()`         | Called via `inspect sandbox cleanup`                                                                                                    | CLI invoked manual cleanup of resources created by this `SandboxEnvironment`.         |
+| Method                  | Lifecycle                                                                                                                                | Purpose                                                                               |
+|-------------------------|------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------|
+| `default_concurrency()` | Called once to determine the default maximum number of sandboxes to run in parallel. Return `None` for no limit (the default behaviour). |                                                                                       |
+| `task_init()`           | Called once for each unique sandbox environment config before executing the tasks in an `eval()` run.                                    | Expensive initialisation operations (e.g. pulling or building images)                 |
+| `sample_init()`         | Called at the beginning of each `Sample`.                                                                                                | Create `SandboxEnvironment` instances for the sample.                                 |
+| `sample_cleanup()`      | Called at the end of each `Sample`                                                                                                       | Cleanup `SandboxEnvironment` instances for the sample.                                |
+| `task_cleanup()`        | Called once for each unique sandbox environment config after executing the tasks in an `eval()` run.                                     | Last chance handler for any resources not yet cleaned up (see also discussion below). |
+| `cli_cleanup()`         | Called via `inspect sandbox cleanup`                                                                                                     | CLI invoked manual cleanup of resources created by this `SandboxEnvironment`.         |
+| `config_files()`        | Called once to determine the names of ‘default’ config files for this provider (e.g. ‘compose.yaml’).                                    |                                                                                       |
+| `config_deserialize()`  | Called when a custom sandbox config type is read from a log file.                                                                        | Only required if a sandbox supports custom config types.                              |
 
 In the case of parallel execution of a group of tasks within the same
 working directory, the `task_init()` and `task_cleanup()` functions will

diff --git a/human-agent.html b/human-agent.html
@@ -304,6 +304,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">
@@ -1116,7 +1119,7 @@ <h2 class="anchored" data-anchor-id="headless">Headless</h2>
     </div>
   </div>
 </footer>
-<script>var lightboxQuarto = GLightbox({"descPosition":"bottom","selector":".lightbox","closeEffect":"zoom","loop":false,"openEffect":"zoom"});
+<script>var lightboxQuarto = GLightbox({"closeEffect":"zoom","selector":".lightbox","descPosition":"bottom","loop":false,"openEffect":"zoom"});
 window.onload = () => {
   lightboxQuarto.on('slide_before_load', (data) => {
     const { slideIndex, slideNode, slideConfig, player, trigger } = data;

diff --git a/index.html b/index.html
@@ -303,6 +303,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">
@@ -622,6 +625,7 @@ <h2 class="anchored" data-anchor-id="learning-more">Learning More</h2>
 <li><p><a href="./caching.html">Caching</a> explains how to cache model output to reduce the number of API calls made.</p></li>
 <li><p><a href="./multimodal.html">Multimodal</a> describes the APIs available for creating multimodal evaluations (including images, audio, and video).</p></li>
 <li><p><a href="./reasoning.html">Reasoning</a> documents the additional options and data available for reasoning models.</p></li>
+<li><p><a href="./structured.html">Structured Output</a> explains how to constrain model output to a particular JSON schema.</p></li>
 </ul>
 <p>These sections describe how to create agent evaluations with Inspect:</p>
 <ul>
@@ -1174,7 +1178,7 @@ <h2 class="anchored" data-anchor-id="learning-more">Learning More</h2>
     </div>
   </div>
 </footer>
-<script>var lightboxQuarto = GLightbox({"closeEffect":"zoom","descPosition":"bottom","openEffect":"zoom","selector":".lightbox","loop":false});
+<script>var lightboxQuarto = GLightbox({"openEffect":"zoom","descPosition":"bottom","selector":".lightbox","closeEffect":"zoom","loop":false});
 window.onload = () => {
   lightboxQuarto.on('slide_before_load', (data) => {
     const { slideIndex, slideNode, slideConfig, player, trigger } = data;

diff --git a/index.html.md b/index.html.md
@@ -274,6 +274,9 @@ These sections cover how to use varoius language models with Inspect:
 - [Reasoning](reasoning.qmd) documents the additional options and data
   available for reasoning models.
 
+- [Structured Output](structured.qmd) explains how to constrain model
+  output to a particular JSON schema.
+
 These sections describe how to create agent evaluations with Inspect:
 
 - [Agents](agents.qmd) combine planning, memory, and tool usage to

diff --git a/interactivity.html b/interactivity.html
@@ -301,6 +301,9 @@
  <span class="menu-text">Reasoning</span></a>
   </div>
 </li>
+          <li class="sidebar-item">
+ <span class="menu-text">structrued.qmd</span>
+  </li>
       </ul>
   </li>
         <li class="sidebar-item sidebar-item-section">