diff --git a/fr/aria/overview.mdx b/fr/aria/overview.mdx index 50fb9fc37f..4c6b39d7d6 100644 --- a/fr/aria/overview.mdx +++ b/fr/aria/overview.mdx @@ -147,7 +147,7 @@ Voici des exemples de conversations réelles avec ARIA pour illustrer comment vo ARIA exécutant une expérience - ARIA utilise W&B Launch pour exécuter des expériences dans un environnement sandbox. Voir [Activer ARIA pour exécuter des expériences](/fr/aria/autoresearch#run-experiments-with-w\&b-launch) pour plus d’informations. + ARIA utilise W&B Launch pour exécuter des expériences dans un environnement sandbox. Voir [Activer ARIA pour exécuter des expériences](/fr/aria/autoresearch#set-up-wb-launch-with-aria) pour plus d’informations. diff --git a/ja/aria/overview.mdx b/ja/aria/overview.mdx index 53328407e8..e7a1a247d4 100644 --- a/ja/aria/overview.mdx +++ b/ja/aria/overview.mdx @@ -148,7 +148,7 @@ ARIA の処理中に、**Thinking** ステップと **shell** ステップが表 experiment を実行している ARIA - ARIA は W&B Launch を使用して、サンドボックス環境で Experiments を実行します。詳しくは、[ARIA が Experiments を実行できるようにする](/ja/aria/autoresearch#run-experiments-with-w\&b-launch) をご覧ください。 + ARIA は W&B Launch を使用して、サンドボックス環境で Experiments を実行します。詳しくは、[ARIA が Experiments を実行できるようにする](/ja/aria/autoresearch#set-up-wb-launch-with-aria) をご覧ください。 diff --git a/ko/aria/overview.mdx b/ko/aria/overview.mdx index 9981665ee7..107d7c4456 100644 --- a/ko/aria/overview.mdx +++ b/ko/aria/overview.mdx @@ -147,7 +147,7 @@ ARIA가 작업하는 동안 **Thinking** 및 **shell** step이 표시될 수 있 실험을 실행하는 ARIA - ARIA는 W&B Launch를 사용해 샌드박스 환경에서 실험을 실행합니다. 자세한 내용은 [ARIA가 실험을 실행하도록 설정](/ko/aria/autoresearch#run-experiments-with-w\&b-launch)을 참조하세요. + ARIA는 W&B Launch를 사용해 샌드박스 환경에서 실험을 실행합니다. 자세한 내용은 [ARIA가 실험을 실행하도록 설정](/ko/aria/autoresearch#set-up-wb-launch-with-aria)을 참조하세요. diff --git a/models/app/features/panels.mdx b/models/app/features/panels.mdx index 4306f6c4a9..08fa0a1f6e 100644 --- a/models/app/features/panels.mdx +++ b/models/app/features/panels.mdx @@ -7,7 +7,7 @@ keywords: ["workspace panels", "quick add", "full-screen mode", "panel sections" import { AriaChatBubbles } from '/snippets/AriaChatBubbles.jsx'; -Use workspace panel visualizations to explore your [logged data](/models/ref/python/experiments/run.md/#method-runlog) by key, visualize the relationships between hyperparameters and output metrics, and more. This page describes how to choose a workspace mode, add and configure panels, organize them into sections, and share them with collaborators. +Use workspace panel visualizations to explore your [logged data](/models/ref/python/experiments/run#method-run-log) by key, visualize the relationships between hyperparameters and output metrics, and more. This page describes how to choose a workspace mode, add and configure panels, organize them into sections, and share them with collaborators. ## Workspace modes @@ -77,7 +77,7 @@ When you work in full-screen mode, use the following tips: - To return to a panel's workspace from full-screen mode, click the left-pointing arrow at the top of the page. - To navigate through a section's panels without exiting full-screen mode, use either the **Previous** and **Next** buttons below the panel or the left and right arrow keys. - To reclaim more space for the panel, minimize the run selector with **Cmd+.** (macOS) or **Ctrl+.** (Windows or Linux). -- When you view an image from a [media panel](/models/app/features/panels/media) in full-screen mode, keyboard shortcuts can zoom in or out, reset zoom, or zoom to fit. See [Keyboard shortcuts](/models/app/keyboard-shortcuts#media-panels). +- When you view an image from a [media panel](/models/app/features/panels/media) in full-screen mode, keyboard shortcuts can zoom in or out, reset zoom, or zoom to fit. See [Keyboard shortcuts](/models/app/keyboard-shortcuts#media-browser-modal). See [Keyboard shortcuts](/models/app/keyboard-shortcuts) for other full-screen and panel shortcuts. diff --git a/models/app/features/panels/code.mdx b/models/app/features/panels/code.mdx index 19d2886671..c8f057aa39 100644 --- a/models/app/features/panels/code.mdx +++ b/models/app/features/panels/code.mdx @@ -64,7 +64,7 @@ with wandb.init() as run: ) ``` -For more control over the types and locations of source code files that W&B saves, see the [reference docs](/models/ref/python/experiments/run#log_code). +For more control over the types and locations of source code files that W&B saves, see the [reference docs](/models/ref/python/experiments/run#method-run-log_code). ### Automatically capture code diff --git a/models/app/features/panels/line-plot/reference.mdx b/models/app/features/panels/line-plot/reference.mdx index d62455f8ad..e66cf44d42 100644 --- a/models/app/features/panels/line-plot/reference.mdx +++ b/models/app/features/panels/line-plot/reference.mdx @@ -347,7 +347,7 @@ The following table describes available hyperbolic functions: The following are example expressions for line plot axes. These examples are for illustrative purposes. You can use any combination of operators and functions described in the previous sections to create complex expressions that suit your needs. -For the following examples, suppose your [summary metrics](/models/track/log/log-summary#log-summary-metrics) include `accuracy` and `loss` with the following values: +For the following examples, suppose your [summary metrics](/models/track/log/log-summary) include `accuracy` and `loss` with the following values: ```json Summary metrics { @@ -428,7 +428,7 @@ sqrt(${summary:loss}*100)+sqrt(${summary:loss}*100000) ### Multi-metric panel expressions -Use a regular expression to create a single line plot that shows multiple metrics together (including matching metrics logged in the future). For detailed instructions, see [Add a line plot](/models/app/features/panels/line-plot#multi-metric-line-plot). +Use a regular expression to create a single line plot that shows multiple metrics together (including matching metrics logged in the future). For detailed instructions, see [Add a line plot](/models/app/features/panels/line-plot#multi-metric-regular-expressions). For example: - Instead of creating separate panels for each layer's metrics, you can view them together in a single panel. For example, if you log metrics with consistent naming, like `layer_0_loss`, `layer_1_loss`, and `layer_2_loss`, you can use a regex like `layer_\d+_loss` to display all layer losses on one plot. diff --git a/models/app/features/panels/query-panels.mdx b/models/app/features/panels/query-panels.mdx index be5825bef5..f89a836bb8 100644 --- a/models/app/features/panels/query-panels.mdx +++ b/models/app/features/panels/query-panels.mdx @@ -81,7 +81,7 @@ runs.summary["cifar10_sample_table"] Breaking this down: -* `runs` is a variable automatically injected in query panel expressions when the query panel is in a workspace. Its value is the list of runs visible for that workspace. For details about the different attributes available within a run, see [Understanding the different attributes](/models/track/public-api-guide/#understanding-the-different-attributes). +* `runs` is a variable automatically injected in query panel expressions when the query panel is in a workspace. Its value is the list of runs visible for that workspace. For details about the different attributes available within a run, see [Export run data](/models/track/public-api-guide#export-run-data). * `summary` is an op that returns the Summary object for a run. Ops are *mapped*, meaning this op is applied to each run in the list, resulting in a list of Summary objects. * `["cifar10_sample_table"]` is a Pick op (denoted with brackets) with a key of `cifar10_sample_table`. Because Summary objects act like dictionaries or maps, this operation picks that field from each Summary object. diff --git a/models/artifacts.mdx b/models/artifacts.mdx index bcc396e067..f50c02ca3b 100644 --- a/models/artifacts.mdx +++ b/models/artifacts.mdx @@ -59,7 +59,7 @@ See the [track external files](/models/artifacts/track-external-files) page for ## Download an artifact -Indicate the artifact you want to mark as input to your run with the [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#use_artifact) method. +Indicate the artifact you want to mark as input to your run with the [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#method-run-use_artifact) method. Continuing from the previous code snippet, the following code example shows how to use the artifact called `example_artifact` that was created earlier: diff --git a/models/artifacts/artifacts-walkthrough.mdx b/models/artifacts/artifacts-walkthrough.mdx index 2744c1580c..e180469828 100644 --- a/models/artifacts/artifacts-walkthrough.mdx +++ b/models/artifacts/artifacts-walkthrough.mdx @@ -83,8 +83,8 @@ with wandb.init(project="artifacts-example", job_type="upload-dataset") as run: The following code example demonstrates the steps you can take to use an artifact you have logged and saved to the W&B servers. 1. First, initialize a new run object with **`wandb.init()`.** -2. Second, use the run objects [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#use_artifact) method to tell W&B what artifact to use. This returns an artifact object. -3. Third, use the artifacts [`wandb.Artifact.download()`](/models/ref/python/experiments/artifact#download) method to download the contents of the artifact. +2. Second, use the run objects [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#method-run-use_artifact) method to tell W&B what artifact to use. This returns an artifact object. +3. Third, use the artifacts [`wandb.Artifact.download()`](/models/ref/python/experiments/artifact#method-artifact-download) method to download the contents of the artifact. ```python # Create a W&B Run. Here we specify 'training' for 'type' diff --git a/models/artifacts/construct-an-artifact.mdx b/models/artifacts/construct-an-artifact.mdx index f65079a181..6a397e83ae 100644 --- a/models/artifacts/construct-an-artifact.mdx +++ b/models/artifacts/construct-an-artifact.mdx @@ -65,13 +65,13 @@ For more information on available attributes and parameters, see the [`wandb.Art [Add files, directories, external URI references (such as Amazon S3) and more](/models/artifacts/construct-an-artifact#add-files-to-an-artifact) to your artifact object. -To add a single file, use the artifact object's [`Artifact.add_file()`](/models/ref/python/experiments/artifact#add_file) method: +To add a single file, use the artifact object's [`Artifact.add_file()`](/models/ref/python/experiments/artifact#method-artifact-add_file) method: ```python artifact.add_file(local_path="path/to/file.txt", name="artifact-name") ``` -To add a directory, use the [`Artifact.add_dir()`](/models/ref/python/experiments/artifact#add_dir) method: +To add a directory, use the [`Artifact.add_dir()`](/models/ref/python/experiments/artifact#method-artifact-add_dir) method: ```python artifact.add_dir(local_path="path/to/directory", name="artifact-name") @@ -81,7 +81,7 @@ See the next section, [Add files to an artifact](/models/artifacts/construct-an- ### Save your artifact to the W&B server -Save your artifact to the W&B server. Use the run object's [`wandb.Run.log_artifact()`](/models/ref/python/experiments/run#log_artifact) method to save the artifact. +Save your artifact to the W&B server. Use the run object's [`wandb.Run.log_artifact()`](/models/ref/python/experiments/run#method-run-log_artifact) method to save the artifact. ```python with wandb.init(project="", job_type="") as run: @@ -148,7 +148,7 @@ root-directory ### Add a single file -Use [`wandb.Artifact.add_file()`](/models/ref/python/experiments/artifact#method-artifact-add-file) to add a single local file to an artifact. Provide the local path to the file as the `local_path` parameter: +Use [`wandb.Artifact.add_file()`](/models/ref/python/experiments/artifact#method-artifact-add_file) to add a single local file to an artifact. Provide the local path to the file as the `local_path` parameter: ```python import wandb @@ -199,7 +199,7 @@ The following table shows how different API calls produce different artifact con ### Add multiple files -Use the [`wandb.Artifact.add_dir()`](/models/ref/python/experiments/artifact#method-artifact-add-dir) method to add multiple files from a local directory to an artifact. Provide the local path to the directory as the `local_path` parameter. +Use the [`wandb.Artifact.add_dir()`](/models/ref/python/experiments/artifact#method-artifact-add_dir) method to add multiple files from a local directory to an artifact. Provide the local path to the directory as the `local_path` parameter. ```python @@ -223,7 +223,7 @@ The following table show how different API calls produce different artifact cont Artifacts track checksums and other information for reproducibility if the URI has a scheme that the W&B library supports. -Add an external URI reference to an artifact with the [`wandb.Artifact.add_reference()`](/models/ref/python/experiments/artifact#method-artifact-add-reference) method. Replace the `'uri'` string with your own URI. Optionally pass the desired path within the artifact for the name parameter. +Add an external URI reference to an artifact with the [`wandb.Artifact.add_reference()`](/models/ref/python/experiments/artifact#method-artifact-add_reference) method. Replace the `'uri'` string with your own URI. Optionally pass the desired path within the artifact for the name parameter. ```python # Add a URI reference diff --git a/models/artifacts/delete-artifacts.mdx b/models/artifacts/delete-artifacts.mdx index af6c966ae9..be3b867d35 100644 --- a/models/artifacts/delete-artifacts.mdx +++ b/models/artifacts/delete-artifacts.mdx @@ -75,7 +75,7 @@ To delete an artifact version: -Delete an artifact version programmatically with the [wandb.Artifact.delete()](/models/ref/python/experiments/artifact#delete) method. Provide the full name of the artifact. The full name consists of `//:`. Set the `delete_aliases` parameter to `True` to delete the artifact even if it has one or more aliases associated with it. +Delete an artifact version programmatically with the [wandb.Artifact.delete()](/models/ref/python/experiments/artifact#method-artifact-delete) method. Provide the full name of the artifact. The full name consists of `//:`. Set the `delete_aliases` parameter to `True` to delete the artifact even if it has one or more aliases associated with it. ```python import wandb @@ -164,7 +164,7 @@ To delete an artifact collection: -Delete artifact collection programmatically with the [wandb.Artifact.delete()](/models/ref/python/experiments/artifact#delete) method. +Delete artifact collection programmatically with the [wandb.Artifact.delete()](/models/ref/python/experiments/artifact#method-artifact-delete) method. Provide the full path of the artifact collection to `wandb.Api.artifact_collection(name="")`. The full path consists of `//`. diff --git a/models/artifacts/download-and-use-an-artifact.mdx b/models/artifacts/download-and-use-an-artifact.mdx index 19644ce124..1d2b004bd9 100644 --- a/models/artifacts/download-and-use-an-artifact.mdx +++ b/models/artifacts/download-and-use-an-artifact.mdx @@ -21,8 +21,8 @@ References that have schemes that W&B knows how to handle get downloaded just li You can download all files in an artifact or download individual files: -- Use [`Artifact.download()`](/models/ref/python/experiments/artifact#download) to download all files. -- Use [`Artifact.get_entry()`](/models/ref/python/experiments/artifact#get_entry) to retrieve a specific entry, and then call `.download()` to download it. +- Use [`Artifact.download()`](/models/ref/python/experiments/artifact#method-artifact-download) to download all files. +- Use [`Artifact.get_entry()`](/models/ref/python/experiments/artifact#method-artifact-get_entry) to retrieve a specific entry, and then call `.download()` to download it. @@ -143,7 +143,7 @@ with wandb.init(project="project-name", job_type="job-type") as run: ## Construct and use an artifact simultaneously -Simultaneously construct and use an artifact. Create an artifact object and pass it to use_artifact. This creates an artifact in W&B if it does not exist yet. The [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#use_artifact) API is idempotent, so you can call it as many times as you like. +Simultaneously construct and use an artifact. Create an artifact object and pass it to use_artifact. This creates an artifact in W&B if it does not exist yet. The [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#method-run-use_artifact) API is idempotent, so you can call it as many times as you like. ```python import wandb diff --git a/models/artifacts/explore-and-traverse-an-artifact-graph.mdx b/models/artifacts/explore-and-traverse-an-artifact-graph.mdx index eec9a780ca..1a0dc6441b 100644 --- a/models/artifacts/explore-and-traverse-an-artifact-graph.mdx +++ b/models/artifacts/explore-and-traverse-an-artifact-graph.mdx @@ -68,7 +68,7 @@ Hover your mouse over a node and click on the eye icon to hide or show a node in Programmatically navigate a graph using the W&B Python SDK. Use an artifact object's -[`logged_by()`](/models/ref/python/experiments/artifact#method-artifact-logged-by) and [`used_by()`](/models/ref/python/experiments/artifact#method-artifact-used-by) methods to walk the graph: +[`logged_by()`](/models/ref/python/experiments/artifact#method-artifact-logged_by) and [`used_by()`](/models/ref/python/experiments/artifact#method-artifact-used_by) methods to walk the graph: ```python with wandb.init() as run: @@ -90,7 +90,7 @@ To enable lineage graph tracking, you need to mark artifacts as [inputs](/models ### Track the input of a run -Mark an artifact as the input (or dependency) of a run with the [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#method-runuse_artifact) +Mark an artifact as the input (or dependency) of a run with the [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#method-run-use_artifact) method. Specify the name of the artifact and an optional alias to reference a specific version of that artifact. The name of the artifact is in the format `:` or `:`. @@ -108,8 +108,8 @@ with wandb.init(entity="", project="") as run: ### Track the output of a run -Use [`wandb.Run.log_artifact()`](/models/ref/python/experiments/run#log_artifact) to declare an artifact as an output of a run. First, -create an artifact with the [`wandb.Artifact()`](/models/ref/python/experiments/artifact#wandb.Artifact) constructor. Then, log the +Use [`wandb.Run.log_artifact()`](/models/ref/python/experiments/run#method-run-log_artifact) to declare an artifact as an output of a run. First, +create an artifact with the [`wandb.Artifact()`](/models/ref/python/experiments/artifact#class-wandb-artifact) constructor. Then, log the artifact as an output of the run with `wandb.Run.log_artifact()`. Replace values enclosed in angle brackets (`< >`) with your values: diff --git a/models/artifacts/ttl.mdx b/models/artifacts/ttl.mdx index 569d61f10f..0b5f65c540 100644 --- a/models/artifacts/ttl.mdx +++ b/models/artifacts/ttl.mdx @@ -71,7 +71,7 @@ The steps are as follows: 1. [Create an artifact](/models/artifacts/construct-an-artifact/). 2. [Add content to the artifact](/models/artifacts/construct-an-artifact/#add-files-to-an-artifact) such as files, a directory, or a reference. 3. Define a TTL time limit with the [`datetime.timedelta`](https://docs.python.org/3/library/datetime.html) data type that is part of Python's standard library. -4. [Log the artifact](/models/artifacts/construct-an-artifact/#3-save-your-artifact-to-the-wb-server). +4. [Log the artifact](/models/artifacts/construct-an-artifact#save-your-artifact-to-the-wb-server). The following code snippet demonstrates how to create an artifact and set a TTL policy. @@ -100,7 +100,7 @@ When you modify an artifact's TTL, the time the artifact takes to expire is stil 1. [Fetch your artifact](/models/artifacts/download-and-use-an-artifact/). 2. Pass in a time delta to the artifact's `ttl` attribute. -3. Update the artifact with the [`save`](/models/ref/python/experiments/run#save) method. +3. Update the artifact with the [`save`](/models/ref/python/experiments/run#method-run-save) method. The following code snippet shows how to set a TTL policy for an artifact: @@ -178,7 +178,7 @@ Artifacts with TTL turned off will not inherit an artifact collection's TTL. Ref 1. [Fetch your artifact](/models/artifacts/download-and-use-an-artifact/). 2. Set the artifact's `ttl` attribute to `None`. -3. Update the artifact with the [`save`](/models/ref/python/experiments/run#save) method. +3. Update the artifact with the [`save`](/models/ref/python/experiments/run#method-run-save) method. The following code snippet shows how to turn off a TTL policy for an artifact: diff --git a/models/integrations/lightning.mdx b/models/integrations/lightning.mdx index 1c4fb9ba5f..db620a0684 100644 --- a/models/integrations/lightning.mdx +++ b/models/integrations/lightning.mdx @@ -303,7 +303,7 @@ for epoch in range(num_epochs): ## Log the min/max of a metric -Using W&B's [`define_metric`](/models/ref/python/experiments/run#define_metric) function, you can define whether your W&B summary metric displays the min, max, mean, or best value for that metric. If `define_metric` isn't used, the last value logged appears in your summary metrics. For more information, see the [customize logging axes guide](/models/track/log/customize-logging-axes/). +Using W&B's [`define_metric`](/models/ref/python/experiments/run#method-run-define_metric) function, you can define whether your W&B summary metric displays the min, max, mean, or best value for that metric. If `define_metric` isn't used, the last value logged appears in your summary metrics. For more information, see the [customize logging axes guide](/models/track/log/customize-logging-axes/). To track the max validation accuracy in the W&B summary metric, call `wandb.define_metric()` only once, at the beginning of training: diff --git a/models/models_quickstart.mdx b/models/models_quickstart.mdx index db31a0ecdb..8fcb4c45c6 100644 --- a/models/models_quickstart.mdx +++ b/models/models_quickstart.mdx @@ -41,7 +41,7 @@ PROJECT = "my-awesome-project" The following code simulates a basic machine learning workflow: training a model, logging metrics, and saving the model as an artifact. -Use the W&B Python SDK (`wandb.sdk`) to interact with W&B during training. Log the loss using [`wandb.Run.log()`](/models/ref/python/experiments/run/#method-runlog), then save the trained model as an artifact using [`wandb.Artifact`](/models/ref/python/experiments/artifact) before finally adding the model file using [`Artifact.add_file`](/models/ref/python/experiments/artifact#add_file). +Use the W&B Python SDK (`wandb.sdk`) to interact with W&B during training. Log the loss using [`wandb.Run.log()`](/models/ref/python/experiments/run#method-run-log), then save the trained model as an artifact using [`wandb.Artifact`](/models/ref/python/experiments/artifact) before finally adding the model file using [`Artifact.add_file`](/models/ref/python/experiments/artifact#method-artifact-add_file). ```python import random # For simulating data @@ -98,7 +98,7 @@ The key takeaways from the previous code block are: * Use `wandb.Run.log()` to log metrics during training. * Use `wandb.Artifact` to save models (datasets, and so forth) as an artifact to your W&B project. -Now that you have trained a model and saved it as an artifact, you can publish it to a registry in W&B. Use [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run/#method-runuse_artifact) to retrieve the artifact from your project and prepare it for publication in the Model registry. `wandb.Run.use_artifact()` serves two key purposes: +Now that you have trained a model and saved it as an artifact, you can publish it to a registry in W&B. Use [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#method-run-use_artifact) to retrieve the artifact from your project and prepare it for publication in the Model registry. `wandb.Run.use_artifact()` serves two key purposes: * Retrieves the artifact object from your project. * Marks the artifact as an input to the run, ensuring reproducibility and traceability. See [Create and view lineage map](/models/registry/lineage) for details. @@ -142,7 +142,7 @@ For additional information on how to link artifacts to a registry, see [Link art ## Retrieve model artifact from registry for inference -To use a model for inference, use `wandb.Run.use_artifact()` to retrieve the published artifact from the registry. This returns an artifact object that you can then use [`wandb.Artifact.download()`](/models/ref/python/experiments/artifact/#method-artifactdownload) to download the artifact to a local file. +To use a model for inference, use `wandb.Run.use_artifact()` to retrieve the published artifact from the registry. This returns an artifact object that you can then use [`wandb.Artifact.download()`](/models/ref/python/experiments/artifact#method-artifact-download) to download the artifact to a local file. ```python REGISTRY_NAME = "Model" # Name of the registry in W&B @@ -252,4 +252,4 @@ for art in artifacts: print(f"artifact created at: {art.created_at}\n") ``` -For more information on querying the registry, see the [Query registry items](/models/registry/search_registry/#query-registry-items-with-mongodb-style-queries). +For more information on querying the registry, see the [Query registry items](/models/registry/search_registry#query-registry-items). diff --git a/models/registry/configure_registry.mdx b/models/registry/configure_registry.mdx index 311486aef2..c93ae994fd 100644 --- a/models/registry/configure_registry.mdx +++ b/models/registry/configure_registry.mdx @@ -3,7 +3,7 @@ title: Configure registry access description: "Configure W&B Registry access by managing users and teams, assigning roles, and setting role-based permissions." --- -A registry admin can [configure registry roles](/models/registry/configure_registry/#configure-registry-roles), [add users](/models/registry/configure_registry/#add-a-user-or-a-team-to-a-registry), or [remove users](/models/registry/configure_registry/#remove-a-user-or-team-from-a-registry) from a registry by configuring the registry's settings. +A registry admin can [configure registry roles](/models/registry/configure_registry/#configure-registry-roles), [add users](/models/registry/configure_registry/#add-a-user-or-a-team), or [remove users](/models/registry/configure_registry/#remove-a-user-or-team) from a registry by configuring the registry's settings. ## Manage users @@ -157,7 +157,7 @@ Service accounts do not inherit elevated registry permissions from their teams. Service accounts are an exception to inherited registry roles. A service account does not inherit elevated registry permissions from its team. For example, if a team has **Admin** access to a registry, service accounts on that team receive only the automatic service account access level, not **Admin** access. -A registry admin can explicitly grant a service account higher access by [adding the service account to the registry with a Member or Admin role](/models/registry/configure_registry/#add-a-user-or-a-team-to-a-registry). +A registry admin can explicitly grant a service account higher access by [adding the service account to the registry with a Member or Admin role](/models/registry/configure_registry/#add-a-user-or-a-team). Automatic service account access depends on registry visibility: @@ -188,16 +188,16 @@ To use the W&B SDK to access artifacts as a **Restricted Viewer**, you must use When a **Restricted Viewer** uses the SDK, certain functions are not available or work differently. The following methods are not available and result in permission errors: -- [`Run.use_artifact()`](/models/ref/python/experiments/run/#method-runuse_artifact) -- [`Artifact.download()`](/models/ref/python/experiments/artifact/#method-artifactdownload) -- [`Artifact.file()`](/models/ref/python/experiments/artifact/#method-artifactfile) -- [`Artifact.files()`](/models/ref/python/experiments/artifact/#method-artifactfiles) +- [`Run.use_artifact()`](/models/ref/python/experiments/run#method-run-use_artifact) +- [`Artifact.download()`](/models/ref/python/experiments/artifact#method-artifact-download) +- [`Artifact.file()`](/models/ref/python/experiments/artifact#method-artifact-file) +- [`Artifact.files()`](/models/ref/python/experiments/artifact#method-artifact-files) The following methods are limited to artifact metadata: -- [`Artifact.get_entry()`](/models/ref/python/experiments/artifact/#method-artifactget_entry) -- [`Artifact.get_path()`](/models/ref/python/experiments/artifact/#method-artifactget_path) -- [`Artifact.get()`](/models/ref/python/experiments/artifact/#method-artifactget) -- [`Artifact.verify()`](/models/ref/python/experiments/artifact/#method-artifactverify) +- [`Artifact.get_entry()`](/models/ref/python/experiments/artifact#method-artifact-get_entry) +- [`Artifact.get_path()`](/models/ref/python/experiments/artifact#method-artifact-get_path) +- [`Artifact.get()`](/models/ref/python/experiments/artifact#method-artifact-get) +- [`Artifact.verify()`](/models/ref/python/experiments/artifact#method-artifact-verify) ### Cross-registry permissions diff --git a/models/registry/create_registry.mdx b/models/registry/create_registry.mdx index 8e6724cf95..ec24aea475 100644 --- a/models/registry/create_registry.mdx +++ b/models/registry/create_registry.mdx @@ -22,7 +22,7 @@ Create a registry either programmatically using the W&B Registry UI or the W&B P 8. Click on the **Create registry** button. -Use the [`wandb.Api().create_registry()`](/models/ref/python/#method-apicreate_registry) method to create a registry programmatically. Provide a name and [visibility](#visibility-types) for the registry for the `name` and `visibility` parameters, respectively. +Use the [`wandb.Api().create_registry()`](/models/ref/python/public-api/api#method-api-create_registry) method to create a registry programmatically. Provide a name and [visibility](#visibility-types) for the registry for the `name` and `visibility` parameters, respectively. Copy and paste the code block below. Replace the values enclosed in `<>` with your own: @@ -35,7 +35,7 @@ registry = wandb.Api().create_registry( ) ``` -See the [`wandb.Api().create_registry()`](/models/ref/python/#method-apicreate_registry) method reference for a full list of parameters that you can provide when you create a registry. +See the [`wandb.Api().create_registry()`](/models/ref/python/public-api/api#method-api-create_registry) method reference for a full list of parameters that you can provide when you create a registry. diff --git a/models/registry/download_use_artifact.mdx b/models/registry/download_use_artifact.mdx index 1290b7946e..3c0978885a 100644 --- a/models/registry/download_use_artifact.mdx +++ b/models/registry/download_use_artifact.mdx @@ -88,7 +88,7 @@ with wandb.init(entity=TEAM_ENTITY, project = PROJECT_NAME) as run: -See [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#use_artifact) and [`Artifact.download()`](/models/ref/python/experiments/artifact#download) in the API Reference for parameters and return type. +See [`wandb.Run.use_artifact()`](/models/ref/python/experiments/run#method-run-use_artifact) and [`Artifact.download()`](/models/ref/python/experiments/artifact#method-artifact-download) in the API Reference for parameters and return type. **Users with a personal entity that belong to multiple organizations** diff --git a/models/registry/link_version.mdx b/models/registry/link_version.mdx index ee1067f26c..4dc887d4b4 100644 --- a/models/registry/link_version.mdx +++ b/models/registry/link_version.mdx @@ -22,8 +22,8 @@ Before you start, check the following: -Programmatically link an artifact version to a collection with [`wandb.Run.link_artifact()`](/models/ref/python/experiments/run#link_artifact) or -[`wandb.Artifact.link()`](/models/ref/python/experiments/artifact#method-artifactlink) +Programmatically link an artifact version to a collection with [`wandb.Run.link_artifact()`](/models/ref/python/experiments/run#method-run-link_artifact) or +[`wandb.Artifact.link()`](/models/ref/python/experiments/artifact#method-artifact-link) The following table summarizes the methods you can use to link an artifact version to a collection and what W&B records when you use each method. @@ -193,7 +193,7 @@ To unlink an artifact version with the W&B Python SDK: 1. Create a Public API client with `wandb.Api()`. 2. Retrieve the linked artifact version with `wandb.Api.artifact()`. -3. Call [`wandb.Artifact.unlink()`](/models/ref/python/experiments/artifact#method-artifactunlink) to unlink the artifact version from the collection. +3. Call [`wandb.Artifact.unlink()`](/models/ref/python/experiments/artifact#method-artifact-unlink) to unlink the artifact version from the collection. ```text wandb-registry-/: diff --git a/models/registry/registry_cards.mdx b/models/registry/registry_cards.mdx index a204841e54..b4157d3ad1 100644 --- a/models/registry/registry_cards.mdx +++ b/models/registry/registry_cards.mdx @@ -51,7 +51,7 @@ Format text with [Markdown markup language](https://www.markdownguide.org/). Use -Use the [`wandb.Api().artifact_collection()`](/models/ref/python/public-api/api#artifact_collection) method to access a collection's description. Use the returned object's `description` property to add, or update, a description to the collection. +Use the [`wandb.Api().artifact_collection()`](/models/ref/python/public-api/api#method-api-artifact_collection) method to access a collection's description. Use the returned object's `description` property to add, or update, a description to the collection. Specify the collection's type for the `type_name` parameter and the collection's full name for the `name` parameter. A collection's name consists of the prefix “wandb-registry”, the name of the registry, and the name of the collection separated by forward slashes: diff --git a/models/registry/search_registry.mdx b/models/registry/search_registry.mdx index 09fff0b4ad..fe08e0f90a 100644 --- a/models/registry/search_registry.mdx +++ b/models/registry/search_registry.mdx @@ -39,7 +39,7 @@ Use the advanced search to search for registries, collections, or artifact versi ## Query registry items -Use [`wandb.Api().registries()`](/models/ref/python/public-api/api#registries) and *query predicates* to filter registries, collections, and artifact versions. A query predicate is a condition that specifies the criteria that returned items must meet. +Use [`wandb.Api().registries()`](/models/ref/python/public-api/api#method-api-registries) and *query predicates* to filter registries, collections, and artifact versions. A query predicate is a condition that specifies the criteria that returned items must meet. To create a query predicate, use a JSON-like dictionary that consists of [query name](/models/registry/search_registry#filterable-fields), one or more [operators](/models/registry/search_registry#supported-operators), and values. The following code snippet shows the general structure of a query predicate: diff --git a/models/reports/create-a-report.mdx b/models/reports/create-a-report.mdx index 97b4b38041..4c9f6c12c3 100644 --- a/models/reports/create-a-report.mdx +++ b/models/reports/create-a-report.mdx @@ -52,7 +52,7 @@ Create a report programmatically: import wandb_workspaces.reports.v2 as wr ``` -3. Create a report instance with `wandb_workspaces.reports.v2.Report` or with the Report Class Public API ([`wandb.apis.reports`](/models/ref/python/public-api/api#reports)). Specify the project to associate with the report: +3. Create a report instance with `wandb_workspaces.reports.v2.Report` or with the Report Class Public API ([`wandb.apis.reports`](/models/ref/python/public-api/api#method-api-reports)). Specify the project to associate with the report: ```python report = wr.Report(project="report_standard") ``` diff --git a/models/runs.mdx b/models/runs.mdx index 2eb0ccf603..3d058a735e 100644 --- a/models/runs.mdx +++ b/models/runs.mdx @@ -15,7 +15,7 @@ Common use cases for initializing and logging to a run include: To initialize a W&B run, call the [`wandb.init()`](/models/ref/python/functions/init) method from the W&B Python SDK. This starts a new run and returns a `wandb.Run` object that you can use to log metrics, artifacts, and other information to the run. For more information about initializing a run, see [Initialize runs](/models/runs/initialize-run). -Each run object has a [unique identifier known as a *run ID*](/models/runs/run-identifiers#unique-run-identifiers). [You can specify a unique ID](/models/runs/run-identifiers#unique-run-identifiers) or let [W&B randomly generate one for you](/models/runs/run-identifiers#autogenerated-run-ids). Each run object also has a human-readable, non-unique [run name](/models/runs/run-identifiers#run-name). You can specify a name for your run or let W&B randomly generate one for you. You can rename a run after initializing it. +Each run object has a [unique identifier known as a *run ID*](/models/runs/run-identifiers#run-id). [You can specify a unique ID](/models/runs/run-identifiers#create-a-custom-run-id) or let [W&B randomly generate one for you](/models/runs/run-identifiers#run-id). Each run object also has a human-readable, non-unique [run name](/models/runs/run-identifiers#run-name). You can specify a name for your run or let W&B randomly generate one for you. You can rename a run after initializing it. W&B logs your run to a [*project*](/models/track/project-page/). You specify the project when you initialize the run with `wandb.init(project="")`. W&B creates a new project if the project does not exist. If the project does exist, W&B logs the run to the project you specified. diff --git a/models/runs/alert.mdx b/models/runs/alert.mdx index 4a00125103..defb9d5a4e 100644 --- a/models/runs/alert.mdx +++ b/models/runs/alert.mdx @@ -33,7 +33,7 @@ If you're using [W&B Server](/platform/hosting/) in your Private Cloud or on W&B W&B can notify you when a run finishes or crashes, without any code changes. In your [User Settings](https://wandb.ai/settings), scroll to the **Alerts** section and turn on: - **Run finished** to be notified when a run completes successfully. -- **Run crashed** to be notified when a run exits without calling [`run.finish()`](/models/ref/python/experiments/run#finish). +- **Run crashed** to be notified when a run exits without calling [`run.finish()`](/models/ref/python/experiments/run#method-run-finish). These toggles are per-user. **Run finished** alerts do not fire for runs in Jupyter notebooks. In notebooks, use `run.alert()` for completion notifications. diff --git a/models/runs/forking.mdx b/models/runs/forking.mdx index 60518e25c9..df9d6fb166 100644 --- a/models/runs/forking.mdx +++ b/models/runs/forking.mdx @@ -11,7 +11,7 @@ Preview features are available for production evaluation, but functionality, API You can explore different hyperparameters or models from a specific point in an experiment without impacting the original run. To do this, fork from an existing W&B run. -When you fork from a run, W&B creates a new run using the source run’s [unique ID](/models/runs/run-identifiers#unique-run-identifiers) and a specified step. Summary metrics from the source run are copied to the forked run. The forked run shares all history and files from the source run up to the specified step. +When you fork from a run, W&B creates a new run using the source run’s [unique ID](/models/runs/run-identifiers#run-id) and a specified step. Summary metrics from the source run are copied to the forked run. The forked run shares all history and files from the source run up to the specified step. After the fork step, you can log new data to the forked run independently of the original run. @@ -20,7 +20,7 @@ View a [live demo](https://wandb.ai/wandb/test-fork-run/workspace?nw=nwuserjulia * Forking a run requires [`wandb`](https://pypi.org/project/wandb/) SDK version >= 0.16.5 -* Forking a run requires monotonically increasing steps. You cannot fork from a run that uses non-monotonic steps defined with [`define_metric()`](/models/ref/python/experiments/run#define_metric). Non-monotonic steps break the chronological order of run history and system metrics. +* Forking a run requires monotonically increasing steps. You cannot fork from a run that uses non-monotonic steps defined with [`define_metric()`](/models/ref/python/experiments/run#method-run-define_metric). Non-monotonic steps break the chronological order of run history and system metrics. Specify the source run's unique `run ID` and the `step` you want to start the forked run from as arguments to `fork_from` in [`wandb.init()`](/models/ref/python/functions/init). diff --git a/models/runs/grouping.mdx b/models/runs/grouping.mdx index fe8e3d539f..c0558c4816 100644 --- a/models/runs/grouping.mdx +++ b/models/runs/grouping.mdx @@ -17,7 +17,7 @@ Use `WANDB_RUN_GROUP` to specify a group for your runs as an environment variabl You can add runs to a group programmatically using the W&B Python SDK or interactively in the W&B App. -W&B stores group names as a run [`wandb.Run.group`](/models/ref/python/experiments/run#property-run-group) property. +W&B stores group names as a run [`wandb.Run.group`](/models/ref/python/experiments/run#properties) property. @@ -96,7 +96,7 @@ To delete a group, remove all runs from it. This automatically deletes the group Organize runs by their *job type*. A job type indicates the function of a run, such as `preprocessing`, `training`, or `evaluation`. -View a run's job type by accessing the run's [`wandb.Run.job_type`](/models/ref/python/experiments/run#property-run-job-type) property. +View a run's job type by accessing the run's [`wandb.Run.job_type`](/models/ref/python/experiments/run#properties) property. Add a job type to a run by passing the `job_type` parameter to `wandb.init(job_type="")`. For example, the following code snippet creates runs with job types of either `training` or `evaluation`: diff --git a/models/runs/manage-runs.mdx b/models/runs/manage-runs.mdx index e01c4fb2eb..9516c13f74 100644 --- a/models/runs/manage-runs.mdx +++ b/models/runs/manage-runs.mdx @@ -16,7 +16,7 @@ To move runs from one project to another or between teams: 5. Select the destination team and project from the dropdown. -When you move a run, historical artifacts associated with it are not moved. To move an artifact manually, you can use the [`wandb artifact get`](/models/ref/cli/wandb-artifact/wandb-artifact-get/) SDK command or the [`Api.artifact` API](/models/ref/python/public-api/api/#artifact) to download the artifact, then use [`wandb artifact put`](/models/ref/cli/wandb-artifact/wandb-artifact-put/) or the `Api.artifact` API to upload it to the run's new location. +When you move a run, historical artifacts associated with it are not moved. To move an artifact manually, you can use the [`wandb artifact get`](/models/ref/cli/wandb-artifact/wandb-artifact-get/) SDK command or the [`Api.artifact` API](/models/ref/python/public-api/api#method-api-artifact) to download the artifact, then use [`wandb artifact put`](/models/ref/cli/wandb-artifact/wandb-artifact-put/) or the `Api.artifact` API to upload it to the run's new location. {/* ## Move runs to a group diff --git a/models/runs/resuming.mdx b/models/runs/resuming.mdx index 9d30cb6a55..45d52fe142 100644 --- a/models/runs/resuming.mdx +++ b/models/runs/resuming.mdx @@ -116,7 +116,7 @@ The instructions below apply when you start sweep agents with the [`wandb agent` Register a handler for the signal that your scheduler or platform uses to indicate preemption, such as `SIGUSR1` or `SIGTERM`. In the handler: -1. Call [`mark_preempting()`](/models/ref/python/experiments/run#mark_preempting) when a run is active. +1. Call [`mark_preempting()`](/models/ref/python/experiments/run#method-run-mark_preempting) when a run is active. 2. Perform any required cleanup, such as saving a checkpoint. 3. Exit with a nonzero status code. A common convention for signal termination is `128 + signum`. diff --git a/models/runs/rewind.mdx b/models/runs/rewind.mdx index 40938755e6..1f4f4bc191 100644 --- a/models/runs/rewind.mdx +++ b/models/runs/rewind.mdx @@ -15,7 +15,7 @@ Due to known performance limitations with rewind, W&B typically recommends [Fork Before you rewind a run, ensure you meet the following prerequisites: * To rewind a run, you must have [W&B Python SDK](https://pypi.org/project/wandb/) version >= `0.17.1`. -* You must use monotonically increasing steps. This does not work with non-monotonic steps defined with [`define_metric()`](/models/ref/python/experiments/run#define_metric) because it disrupts the required chronological order of run history. +* You must use monotonically increasing steps. This does not work with non-monotonic steps defined with [`define_metric()`](/models/ref/python/experiments/run#method-run-define_metric) because it disrupts the required chronological order of run history. ### Limitations Rewind does not support the following: diff --git a/models/runs/run-identifiers.mdx b/models/runs/run-identifiers.mdx index bd4f4cbc01..0a88d6b6fc 100644 --- a/models/runs/run-identifiers.mdx +++ b/models/runs/run-identifiers.mdx @@ -121,7 +121,7 @@ Rename a run after initializing it programmatically with the Python SDK or inter -Use [`wandb.Api.Run`](/models/ref/python/public-api/api#method-api-run) to access a run logged to W&B. This method returns a [run object](/models/ref/python/public-api/run#property-run-name) that you can use to update the run name. Call `wandb.Api.Run.update()` method to persist changes. +Use [`wandb.Api.Run`](/models/ref/python/public-api/api#method-api-run) to access a run logged to W&B. This method returns a [run object](/models/ref/python/public-api/run#properties) that you can use to update the run name. Call `wandb.Api.Run.update()` method to persist changes. Replace the values enclosed in angle brackets (`< >`) with your own values. diff --git a/models/sweeps/signal-handling-sweep-runs.mdx b/models/sweeps/signal-handling-sweep-runs.mdx index 45b666004e..55885f6e58 100644 --- a/models/sweeps/signal-handling-sweep-runs.mdx +++ b/models/sweeps/signal-handling-sweep-runs.mdx @@ -12,11 +12,11 @@ W&B uses the training process exit status to decide whether a run is requeued an **Exit code contract:** - **Exit code 0**: W&B considers the run to have completed successfully and doesn't requeue it. -- **Non-zero exit code**: W&B treats the run as failed or preempted. When you use [`mark_preempting()`](/models/ref/python/experiments/run#mark_preempting), W&B requeues the run so another agent (or the same agent after restart) can resume it. +- **Non-zero exit code**: W&B treats the run as failed or preempted. When you use [`mark_preempting()`](/models/ref/python/experiments/run#method-run-mark_preempting), W&B requeues the run so another agent (or the same agent after restart) can resume it. This applies whether the process exits from a signal handler, from an exception, or from an explicit `sys.exit()` call. Understanding and relying on this contract matters in preemptible or cluster environments. -When the process exits due to a [catchable signal](#catchable-signals-and-preemption), your handler can run, call [`wandb.run.mark_preempting()`](/models/ref/python/experiments/run#mark_preempting) if you want the run requeued, perform cleanup (for example, save a checkpoint), then exit with a non-zero code. A common convention is `sys.exit(128 + signum)` for termination by signal. W&B records that exit code and the same [requeue rules](/models/runs/resuming#resume-preemptible-sweeps-runs) apply. When the operating system kernel kills the process with [`SIGKILL`](#sigkill-uncatchable), the process can't run exit hooks, so W&B doesn't write a final summary and the run might appear as crashed or killed. The agent still starts the next run. +When the process exits due to a [catchable signal](#catchable-signals-and-preemption), your handler can run, call [`wandb.run.mark_preempting()`](/models/ref/python/experiments/run#method-run-mark_preempting) if you want the run requeued, perform cleanup (for example, save a checkpoint), then exit with a non-zero code. A common convention is `sys.exit(128 + signum)` for termination by signal. W&B records that exit code and the same [requeue rules](/models/runs/resuming#resume-preemptible-sweeps-runs) apply. When the operating system kernel kills the process with [`SIGKILL`](#sigkill-uncatchable), the process can't run exit hooks, so W&B doesn't write a final summary and the run might appear as crashed or killed. The agent still starts the next run. ## Stale runs and server-side timeouts @@ -29,7 +29,7 @@ Most signals in preemptible environments are catchable, meaning your training sc **Best practices:** - Register handlers early (for example, before entering the main training loop). -- In the handler, call [`wandb.run.mark_preempting()`](/models/ref/python/experiments/run#mark_preempting) when you intend the run to be requeued after preemption, perform cleanup (for example, save a checkpoint), then exit with a non-zero code. +- In the handler, call [`wandb.run.mark_preempting()`](/models/ref/python/experiments/run#method-run-mark_preempting) when you intend the run to be requeued after preemption, perform cleanup (for example, save a checkpoint), then exit with a non-zero code. The following example registers handlers for `SIGUSR1` (a typical cluster preemption signal) and `SIGTERM`. It leaves `SIGINT` free for interactive use (for example, manual cancellation from the terminal). The handler calls `wandb.run.mark_preempting()` and exits using `128 + signum`: @@ -78,7 +78,7 @@ wandb agent --forward-signals entity/project/sweep_ID W&B doesn't support signal forwarding for [`wandb.agent()`](/models/ref/python/functions/agent) in the Python API. That path runs your training function in a thread, not as a separate child process, so the same forwarding behavior doesn't apply. -When the CLI agent receives `SIGINT` or `SIGTERM` with forwarding enabled, it relays the signal to the child. Your training script's handler can then run, call `wandb.run.mark_preempting()` and [`wandb.finish()`](/models/ref/python/experiments/run#finish) with a non-zero exit code if needed, and exit with a non-zero code. If you press Ctrl+C twice on the agent process, the agent receives `SIGTERM` by default. With `--forward-signals`, the agent can forward `SIGINT` to the child so your handler runs. +When the CLI agent receives `SIGINT` or `SIGTERM` with forwarding enabled, it relays the signal to the child. Your training script's handler can then run, call `wandb.run.mark_preempting()` and [`wandb.finish()`](/models/ref/python/experiments/run#method-run-finish) with a non-zero exit code if needed, and exit with a non-zero code. If you press Ctrl+C twice on the agent process, the agent receives `SIGTERM` by default. With `--forward-signals`, the agent can forward `SIGINT` to the child so your handler runs. For more information, see the [`wandb agent`](/models/ref/cli/wandb-agent) CLI reference. @@ -88,7 +88,7 @@ This section describes how to configure sweeps so that runs survive preemption o **Ensure the training process receives the signal:** -- **When the scheduler signals the agent**: Run the agent with `wandb agent --forward-signals` so that when the scheduler (or user) sends a signal to the agent, the agent forwards it to the child. The child's handler can then call `wandb.run.mark_preempting()`, [`wandb.finish(exit_code=...)`](/models/ref/python/experiments/run#finish) with a non-zero code, and `sys.exit(128 + signum)` (or another non-zero exit code). +- **When the scheduler signals the agent**: Run the agent with `wandb agent --forward-signals` so that when the scheduler (or user) sends a signal to the agent, the agent forwards it to the child. The child's handler can then call `wandb.run.mark_preempting()`, [`wandb.finish(exit_code=...)`](/models/ref/python/experiments/run#method-run-finish) with a non-zero code, and `sys.exit(128 + signum)` (or another non-zero exit code). - **When the scheduler signals the launch script (not the agent directly)**: Have the launch script send the preemption signal directly to the training process. For example, the training script writes its process ID to a file. The launch script traps the cluster signal (for example, `SIGUSR1`) and runs `kill -SIGUSR1 $(cat $PID_FILE)` so the training process's handler runs. **In the training script:** Register a handler for the signal your cluster uses (for example, `SIGTERM` or `SIGUSR1`). In the handler, call `wandb.run.mark_preempting()` if a run is active, then finish the run with a non-zero exit code and `sys.exit(128 + signum)` (or another non-zero code) so W&B requeues the run. For more information about when W&B requeues runs and how that interacts with `mark_preempting()`, see [Resume preemptible Sweeps runs](/models/runs/resuming#resume-preemptible-sweeps-runs). diff --git a/models/track/config.mdx b/models/track/config.mdx index 754e50fdf1..b246292f10 100644 --- a/models/track/config.mdx +++ b/models/track/config.mdx @@ -263,7 +263,7 @@ The raw JSON format is useful when you write expressions to create or transform -Use the [`wandb.Run.config`](/models/ref/python/experiments/run#property-config) property to access a run's configuration values during a run. `wandb.Run.config` returns a dictionary object that contains the configuration key-value pairs. +Use the [`wandb.Run.config`](/models/ref/python/experiments/run#properties) property to access a run's configuration values during a run. `wandb.Run.config` returns a dictionary object that contains the configuration key-value pairs. The following example shows how to access a run's configuration values during a run. Replace the values in the config dictionary with your own hyperparameters, model architecture, dataset name, and other settings. Replace `""` with your W&B project name: diff --git a/models/track/create-an-experiment.mdx b/models/track/create-an-experiment.mdx index 96346076b0..25950b4f0c 100644 --- a/models/track/create-an-experiment.mdx +++ b/models/track/create-an-experiment.mdx @@ -52,7 +52,7 @@ with wandb.init( For more information on how to configure an experiment, see [Configure Experiments](./config). ### Log metrics inside your training loop -Call [`run.log()`](/models/ref/python/experiments/run/#method-runlog) to log metrics about each training step such as accuracy and loss. +Call [`run.log()`](/models/ref/python/experiments/run#method-run-log) to log metrics about each training step such as accuracy and loss. ```python model, dataloader = get_model(), get_data() @@ -125,7 +125,7 @@ The following are some suggested guidelines to consider when you create experime # In a different cell: run.finish() ``` -- **Config**: Track hyperparameters, model architecture, dataset information, and other values needed to reproduce your model. For more information, see [View the config in the Overview section of a run in the W&B App](/models/track/config#view-config-values-in-the-w&b-app). +- **Config**: Track hyperparameters, model architecture, dataset information, and other values needed to reproduce your model. For more information, see [View the config in the Overview section of a run in the W&B App](/models/track/config#view-config-values). - **Project**: Use [projects](/models/track/project-page) to organize experiments in a central location where you can visualize results, compare runs, view and download artifacts, create automations, and more. - **Notes**: Add notes to describe the purpose of a run, such as `baseline model` or `tuned hyperparameters`. You can edit notes later from the run overview in the W&B App. - **Job types**: [Add job types to your runs](/models/runs/grouping#organize-runs-by-job-type) to organize and filter runs by task, such as `train`, `test`, or `inference`. diff --git a/models/track/limits.mdx b/models/track/limits.mdx index 2df69b952f..1c2965feb5 100644 --- a/models/track/limits.mdx +++ b/models/track/limits.mdx @@ -299,7 +299,7 @@ By default, a workspace in automatic mode creates standard panels for each logge To improve performance: 1. Reset the workspace to manual mode. -2. Use [Quick add](/models/app/features/panels/#quick-add) to add only the panels you need. +2. Use [Quick panel builder](/models/app/features/panels#quick-panel-builder) to add only the panels you need. Deleting unused panels one at a time usually has little effect. Reset the workspace and add back only the panels you want. diff --git a/models/track/log.mdx b/models/track/log.mdx index ecb302c8fd..20211bfaf8 100644 --- a/models/track/log.mdx +++ b/models/track/log.mdx @@ -90,7 +90,7 @@ Avoid naming metrics with invalid characters (such as commas, spaces, or special with wandb.init() as run: run.log({'acc': 0.9, 'epoch': 3, 'batch': 117}) ``` - To set the default x-axis for a given metric use [Run.define_metric()](/models/ref/python/experiments/run#define_metric). + To set the default x-axis for a given metric use [Run.define_metric()](/models/ref/python/experiments/run#method-run-define_metric). 4. **Log rich media and charts**: `wandb.Run.log()` supports the logging of a wide variety of data types, from [media like images and videos](/models/track/log/media/) to [tables](/models/track/log/log-tables/) and [charts](/models/app/features/custom-charts/). diff --git a/models/track/log/customize-logging-axes.mdx b/models/track/log/customize-logging-axes.mdx index f10cc783e3..b9d3174fe0 100644 --- a/models/track/log/customize-logging-axes.mdx +++ b/models/track/log/customize-logging-axes.mdx @@ -25,7 +25,7 @@ In the project's workspace, the `validation_loss` metric is plotted against the Line plot panel that uses `step` as the x-axis. -In certain situations, it makes more sense to log metrics against a different x-axis such as a logarithmic x-axis. Use the [`define_metric()`](/models/ref/python/experiments/run/#define_metric) method to use any metric you log as a custom x-axis. +In certain situations, it makes more sense to log metrics against a different x-axis such as a logarithmic x-axis. Use the [`define_metric()`](/models/ref/python/experiments/run#method-run-define_metric) method to use any metric you log as a custom x-axis. Specify the metric that you want to appear as the y-axis with the `name` parameter. The `step_metric` parameter specifies the metric you want to use as the x-axis. When you log a custom metric, specify a value for both the x-axis and the y-axis as key-value pairs in a dictionary. diff --git a/models/track/log/distributed-training.mdx b/models/track/log/distributed-training.mdx index 1f57fd3b55..531eb50b6a 100644 --- a/models/track/log/distributed-training.mdx +++ b/models/track/log/distributed-training.mdx @@ -30,7 +30,7 @@ Explore the code behind these examples in the W&B GitHub examples repository [he This section describes how to track values and metrics available to your rank 0 process. Use this approach to track only metrics that are available from a single process. Typical metrics include GPU/CPU utilization, behavior on a shared validation set, gradients and parameters, and loss values on representative data examples. -Within the rank 0 process, initialize a W&B run with [`wandb.init()`](/models/ref/python/functions/init) and log experiments ([`wandb.Run.log()`](/models/ref/python/experiments/run/#method-runlog)) to that run. +Within the rank 0 process, initialize a W&B run with [`wandb.init()`](/models/ref/python/functions/init) and log experiments ([`wandb.Run.log()`](/models/ref/python/experiments/run#method-run-log)) to that run. The following [sample Python script (`log-ddp.py`)](https://github.com/wandb/examples/blob/master/examples/pytorch/pytorch-ddp/log-ddp.py) demonstrates one way to track metrics on two GPUs on a single machine using PyTorch DDP. [PyTorch DDP](https://pytorch.org/tutorials/intermediate/ddp_tutorial.html) (`DistributedDataParallel` in`torch.nn`) is a popular library for distributed training. The basic principles apply to any distributed training setup, but the implementation may differ. diff --git a/models/track/log/log-models.mdx b/models/track/log/log-models.mdx index 326ce381ba..648fad961b 100644 --- a/models/track/log/log-models.mdx +++ b/models/track/log/log-models.mdx @@ -22,11 +22,11 @@ For more information on W&B Artifacts and advanced versioning use cases, see the ## Log a model to a run -Use the [`log_model`](/models/ref/python/experiments/run#log_model) to log a model artifact that contains content within a directory you specify. The [`log_model`](/models/ref/python/experiments/run#log_model) method also marks the resulting model artifact as an output of the W&B run. +Use the [`log_model`](/models/ref/python/experiments/run#method-run-log_model) to log a model artifact that contains content within a directory you specify. The [`log_model`](/models/ref/python/experiments/run#method-run-log_model) method also marks the resulting model artifact as an output of the W&B run. You can track a model's dependencies and the model's associations if you mark the model as the input or output of a W&B run. View the lineage of the model within the W&B App UI. See the [Explore and traverse artifact graphs](/models/artifacts/explore-and-traverse-an-artifact-graph/) page within the [Artifacts](/models/artifacts/) chapter for more information. -Provide the path where your model files are saved to the `path` parameter. The path can be a local file, directory, or [reference URI](/models/artifacts/track-external-files/#amazon-s3--gcs--azure-blob-storage-references) to an external bucket such as `s3://bucket/path`. +Provide the path where your model files are saved to the `path` parameter. The path can be a local file, directory, or [reference URI](/models/artifacts/track-external-files#track-an-artifact-in-an-external-bucket) to an external bucket such as `s3://bucket/path`. Ensure to replace values enclosed in `<>` with your own. @@ -43,10 +43,10 @@ with wandb.init(project="", entity="") as run: Optionally provide a name for the model artifact for the `name` parameter. If `name` is not specified, W&B will use the basename of the input path prepended with the run ID as the name. -Keep track of the `name` that you, or W&B assigns, to the model. You will need the name of the model to retrieve the model path with the [`wandb.Run.use_model()`](/models/ref/python/experiments/run#use_model) method. +Keep track of the `name` that you, or W&B assigns, to the model. You will need the name of the model to retrieve the model path with the [`wandb.Run.use_model()`](/models/ref/python/experiments/run#method-run-use_model) method. -See [`log_model`](/models/ref/python/experiments/run#log_model) in the API Reference for parameters. +See [`log_model`](/models/ref/python/experiments/run#method-run-log_model) in the API Reference for parameters.
@@ -109,7 +109,7 @@ Find logs at: ./wandb/run-20231206_103511-wlby6fuw/logs ## Download and use a logged model -Use the [`use_model`](/models/ref/python/experiments/run#use_model) function to access and download models files previously logged to a W&B run. +Use the [`use_model`](/models/ref/python/experiments/run#method-run-use_model) function to access and download models files previously logged to a W&B run. Provide the name of the model artifact where the model files you want to retrieve are stored. The name you provide must match the name of an existing logged model artifact. @@ -127,7 +127,7 @@ with wandb.init(project="", entity="") as run: downloaded_model_path = run.use_model(name="") ``` -The [use_model](/models/ref/python/experiments/run#use_model) function returns the path of downloaded model files. Keep track of this path if you want to link this model later. In the preceding code snippet, the returned path is stored in a variable called `downloaded_model_path`. +The [use_model](/models/ref/python/experiments/run#method-run-use_model) function returns the path of downloaded model files. Keep track of this path if you want to link this model later. In the preceding code snippet, the returned path is stored in a variable called `downloaded_model_path`.
@@ -150,4 +150,4 @@ with wandb.init(project=project, entity=entity) as run: ```
-See [`use_model`](/models/ref/python/experiments/run#use_model) in the API Reference for parameters and return type. \ No newline at end of file +See [`use_model`](/models/ref/python/experiments/run#method-run-use_model) in the API Reference for parameters and return type. \ No newline at end of file diff --git a/models/track/log/media.mdx b/models/track/log/media.mdx index 50ce61f9bc..6dc47fe2b5 100644 --- a/models/track/log/media.mdx +++ b/models/track/log/media.mdx @@ -35,7 +35,7 @@ Log images to track inputs, outputs, filter weights, activations, and more. Images can be logged directly from NumPy arrays, as PIL images, or from the filesystem. -Each time you log images from a step, they are available in the UI. Expand the image panel, then use the step slider to look at images from different steps. This makes it easy to compare how a model's output changes during training. Click a media panel to view an image in full-screen mode; there you can zoom and pan, including with [keyboard shortcuts](/models/app/keyboard-shortcuts#media-panels). +Each time you log images from a step, they are available in the UI. Expand the image panel, then use the step slider to look at images from different steps. This makes it easy to compare how a model's output changes during training. Click a media panel to view an image in full-screen mode; there you can zoom and pan, including with [keyboard shortcuts](/models/app/keyboard-shortcuts#media-browser-modal). To compare images or videos from different runs, steps, or indices in one view, use [Compare mode](/models/app/features/panels/media#compare-mode) in a media panel. @@ -346,7 +346,7 @@ Three different formats of NumPy arrays are supported for flexible color schemes #### Python object -Using this schema, you can define a Python object and pass it in to [the `from_point_cloud` method](/models/ref/python/#from_point_cloud). +Using this schema, you can define a Python object and pass it in to [the `from_point_cloud` method](/models/ref/python/data-types/object3d#method-object3d-from_point_cloud). * `points`is a NumPy array containing coordinates and colors for the points to render using [the same formats as the simple point cloud renderer shown above](#python-object). * `boxes` is a NumPy array of python dictionaries with three attributes: @@ -403,7 +403,7 @@ When viewing a point cloud, you can hold control and use the mouse to move aroun #### Point cloud files -You can use [the `from_file` method](/models/ref/python/#from_file) to load in a JSON file full of point cloud data. +You can use [the `from_file` method](/models/ref/python/data-types/object3d#method-object3d-from_file) to load in a JSON file full of point cloud data. ```python run.log({"my_cloud_from_file": wandb.Object3D.from_file( @@ -499,7 +499,7 @@ An example of how to format the point cloud data is shown below. ``` #### NumPy arrays -Using [the same array formats defined above](#numpy-array-formats), you can use `numpy` arrays directly with [the `from_numpy` method](/models/ref/python/#from_numpy) to define a point cloud. +Using [the same array formats defined above](#numpy-array-formats), you can use `numpy` arrays directly with [the `from_numpy` method](/models/ref/python/data-types/object3d#method-object3d-from_numpy) to define a point cloud. ```python run.log({"my_cloud_from_numpy_xyz": wandb.Object3D.from_numpy( diff --git a/models/track/log/working-with-csv.mdx b/models/track/log/working-with-csv.mdx index 44bae53e97..57e779fbab 100644 --- a/models/track/log/working-with-csv.mdx +++ b/models/track/log/working-with-csv.mdx @@ -100,7 +100,7 @@ The below table becomes this W&B Dashboard after conversion In some cases, you might have your experiment details in a CSV file. Common details found in such CSV files include: * A name for the experiment run -* Initial [notes](/models/runs/#add-a-note-to-a-run) +* Initial [notes](/models/runs) * [Tags](/models/runs/tags/) to differentiate the experiments * Configurations needed for your experiment (with the added benefit of being able to utilize our [Sweeps Hyperparameter Tuning](/models/sweeps/)). @@ -160,13 +160,13 @@ for i, row in loaded_experiment_df.iterrows(): ) as run: ``` -As an experiment runs, you might want to log every instance of your metrics so they are available to view, query, and analyze with W&B. Use the [`run.log()`](/models/ref/python/experiments/run/#method-runlog) command to accomplish this: +As an experiment runs, you might want to log every instance of your metrics so they are available to view, query, and analyze with W&B. Use the [`run.log()`](/models/ref/python/experiments/run#method-run-log) command to accomplish this: ```python run.log({key: val}) ``` -You can optionally log a final summary metric to define the outcome of the run using the [`define_metric`](/models/ref/python/experiments/run#define_metric) API. This example adds the summary metrics to our run with `run.summary.update()`: +You can optionally log a final summary metric to define the outcome of the run using the [`define_metric`](/models/ref/python/experiments/run#method-run-define_metric) API. This example adds the summary metrics to our run with `run.summary.update()`: ```python run.summary.update(summaries) diff --git a/platform/launch/create-launch-job.mdx b/platform/launch/create-launch-job.mdx index 9eb2bf32fc..3fa021bcad 100644 --- a/platform/launch/create-launch-job.mdx +++ b/platform/launch/create-launch-job.mdx @@ -97,7 +97,7 @@ wandb launch --docker-image "wandb/job_hello_world:main" --project "hello-world" In addition to jobs you create explicitly with `wandb launch`, W&B can create jobs for you as a side effect of running tracked code. W&B automatically creates and tracks a job for any run with tracked source code, even if that run wasn't created with Launch. Runs are considered to have tracked source code if any of the three following conditions are met: - The run has an associated Git remote and commit hash. -- The run logged a code artifact. See [`Run.log_code`](/models/ref/python/experiments/run#log_code). +- The run logged a code artifact. See [`Run.log_code`](/models/ref/python/experiments/run#method-run-log_code). - The run was executed in a Docker container with the `WANDB_DOCKER` environment variable set to an image tag. If W&B automatically creates your Launch job from a W&B run, W&B infers the Git remote URL from the local Git repository. diff --git a/release-notes/server-releases-archived.mdx b/release-notes/server-releases-archived.mdx index 130276437c..e380671ea6 100644 --- a/release-notes/server-releases-archived.mdx +++ b/release-notes/server-releases-archived.mdx @@ -198,8 +198,8 @@ Private preview features are available by invitation only. To request enrollment - You can now import [human annotations](/weave/guides/tracking/feedback#add-human-annotations) into Weave datasets - [Playground](/weave/guides/tools/playground) now supports saved configurations and prompts. - Decorators are now supported in TypeScript. -- Added support for [tracing generator functions](/weave/guides/tracking/tracing#trace-sync-%26-async-generator-functions). -- The new [`dataset.add_rows`](/weave/reference/python-sdk/#method-add-rows) helper improves the efficiency of appending to an existing dataset. +- Added support for [tracing generator functions](/weave/guides/tracking/tracing). +- The new [`dataset.add_rows`](/weave/reference/python-sdk#method-add_rows) helper improves the efficiency of appending to an existing dataset. - To help you understand your usage, trace and object sizes are now shown through the UI. ## Performance @@ -214,7 +214,7 @@ Private preview features are available by invitation only. To request enrollment - Fixed many cases where y-axes would over-round to a degree that caused duplicate values to display. - Fixed confusing behavior when entering invalid smoothing parameters. - Removed the **Partial Media** warning from media panels. This does not change the behavior of the media panels. -- When adding a [run filter based on tags](/models/runs/filter-runs/#filter-runs-with-tags), the filter is now selected by default, as when filtering by other fields. +- When adding a [run filter based on tags](/models/runs/filter-runs#example-filter-runs-with-tags), the filter is now selected by default, as when filtering by other fields. - Removed the green bell icon that could appear on active runs in the run selector. - Removed the System page for individual runs. - The project description field now respects new lines. @@ -257,10 +257,10 @@ Private preview features are available by invitation only. To request enrollment - **Improved Exponentially-weighted Moving Average (EMA) smoothing** provides more reliable [smoothed lines](/models/app/features/panels/line-plot/smoothing) when operating on complete, unbinned data. In most cases, smoothing is handled at the back end for improved performance. ### Weave -- Chat with fine-tuned models from within your W&B instance. [Playground](/weave/guides/tools/playground) is now supported in Dedicated Cloud. Playground is a chat interface for comparing different LLMs on historical traces. Admins can add API keys to different model providers or hook up [custom hosted LLM providers](/weave/guides/tools/playground/#add-a-custom-provider) so your team can interact with them from within Weave. +- Chat with fine-tuned models from within your W&B instance. [Playground](/weave/guides/tools/playground) is now supported in Dedicated Cloud. Playground is a chat interface for comparing different LLMs on historical traces. Admins can add API keys to different model providers or hook up [custom hosted LLM providers](/weave/guides/tools/playground#add-provider-credentials-and-information) so your team can interact with them from within Weave. - Open Telemetry Support. Now you can log traces via OpenTelemetry (OTel). See [OpenTelemetry tracing](/weave/guides/tracking/otel?utm_source=beamer&utm_medium=sidebar&utm_campaign=OpenTelemetry-support-in-Weave&utm_content=ctalink). - Weave [tracing](/weave/guides/tracking) has new framework integrations: CrewAI, OpenAI's Agent SDK, DSPy 2.x and Google's genai Python SDK. -- Playground supports new [OpenAI models](/weave/guides/tools/playground/#openai): GPT‑4.1, GPT‑4.1 mini, and GPT‑4.1 nano. +- Playground supports new [OpenAI models](/weave/guides/tools/playground#select-an-llm): GPT‑4.1, GPT‑4.1 mini, and GPT‑4.1 nano. - Build labeled datasets directly from traces, with your annotations automatically converted into dataset columns. See [Dataset creation from traces](/weave/guides/core-types/datasets/#create-edit-and-delete-a-dataset-in-the-ui). ## Security diff --git a/release-notes/server-releases.mdx b/release-notes/server-releases.mdx index a9fe79f6bb..99344997f5 100644 --- a/release-notes/server-releases.mdx +++ b/release-notes/server-releases.mdx @@ -141,7 +141,7 @@ W&B Server 0.83.0 makes the Registry cross-platform accessible. On Dedicated Clo In v0.83.x, the following Automations features are available in Dedicated Cloud only: -- **Automation Hub** centralizes the creation and management of [automations](/models/automations#automations-overview) across projects and registries. Find it in the **Core** section of the sidebar. +- **Automation Hub** centralizes the creation and management of [automations](/models/automations) across projects and registries. Find it in the **Core** section of the sidebar. - An automation can now be scoped to an organization, a team, a project, a registry, or a collection. This eliminates the need to duplicate an automation to achieve broad coverage. When creating an automation, click **Edit scopes**. ### Registry @@ -149,7 +149,7 @@ In v0.83.x, the following Automations features are available in Dedicated Cloud In v0.83.x, the following Registry features are available in Dedicated Cloud only: - All Models and Weave users in an organization now have Registry access by default. Existing per-registry permissions are unchanged. Administrators can revoke a user's organization-level Registry access through SCIM. See [Update Registry access](/platform/hosting/iam/scim#update-registry-access). -- Weave [prompts](/weave/guides/core-types/prompts#create-prompt-objects) (`StringPrompt` and `MessagesPrompt`) can now be added to the Registry, so you can version, alias, and reuse prompts across your organization alongside model artifacts. +- Weave [prompts](/weave/guides/core-types/prompts) (`StringPrompt` and `MessagesPrompt`) can now be added to the Registry, so you can version, alias, and reuse prompts across your organization alongside model artifacts. - Improved [artifact collection](/models/registry/create_collection) overview editing and version grid selection. Auto-generate rich collection cards with dynamic artifact panels for more context. ### Weave @@ -159,7 +159,7 @@ In v0.83.x, the following Registry features are available in Dedicated Cloud onl - Agent reactions (thumbs-up/thumbs-down) now save as tags instead of emoji reactions, so they appear alongside scorer tags and you can filter and chart them by agent. - Signals and Playground agents are now hidden by default. - Moved dataset manipulation to the [**Evaluations** page](/weave/guides/core-types/evaluations) and improved the experience of creating datapoints. -- Added an [audio](/weave/cookbooks/audio_with_weave#audio-with-weave) player to agent conversations, including a new **Play all** control that plays each file in sequence. +- Added an [audio](/weave/cookbooks/audio_with_weave) player to agent conversations, including a new **Play all** control that plays each file in sequence. - Improved the [**Agents**](/weave/custom-agents-quickstart) tab. - Added a **Spans** minimap to agent conversations, with turn-anchored feedback data. - Added copy controls to JSON embedded in user messages. @@ -655,7 +655,7 @@ Refer to [Release policies and processes](/release-notes/release-policies). For ![Baseline run with distinct styling in line plot](/images/release-notes/v0-78-0/baseline-line-plot-styling.png) -- [Create panels with multiple metrics](/models/app/features/panels/line-plot#multi-metric-line-plot) in bulk to quickly organize your workspace and spot patterns across large metric sets. For example, you can overlay gradient metrics across model layers or position specific layers side-by-side. [Watch a video demonstration](https://www.loom.com/embed/ebbb1ab3f858473cb24420bf3abf7166). +- [Create panels with multiple metrics](/models/app/features/panels/line-plot#multi-metric-regular-expressions) in bulk to quickly organize your workspace and spot patterns across large metric sets. For example, you can overlay gradient metrics across model layers or position specific layers side-by-side. [Watch a video demonstration](https://www.loom.com/embed/ebbb1ab3f858473cb24420bf3abf7166). - [Workspace templates](/models/app/features/cascade-settings#workspace-layout-options) have new settings. You can now configure your workspace template to hide empty sections, sort panels in a section alphabetically, or adjust how automatic panel sorting in a section works. See [Workspace layout options](/models/app/features/cascade-settings#workspace-layout-options). - The **Sweeps** table now includes a column with the **Sweep ID**. - Admins can now customize usage and spend alerts for your organization. View usage and spending and set up email notifications for storage, tracked hours, Weave, and Training in your organization settings under the new **Usage & Alerts** tab. See [Plan usage](/platform/app/settings-page/billing-settings#plan-usage). @@ -690,7 +690,7 @@ Refer to [Release policies and processes](/release-notes/release-policies). For ![Video: re-ordering columns in Traces and Evaluations tables](/images/release-notes/v0-78-0/weave-column-reorder-demo.gif) - You can now save a customized [Leaderboard](/weave/guides/evaluation/dynamic_leaderboards) in a saved view. - You can now evaluate audio and image content in your traces using [monitors](/weave/guides/evaluation/monitors). -- Aggregated OpenTelemetry token counts now display in the trace detail sidebar for [OpenTelemetry traces](/weave/guides/tracking/otel#send-opentelemetry-traces-to-weave). +- Aggregated OpenTelemetry token counts now display in the trace detail sidebar for [OpenTelemetry traces](/weave/guides/tracking/otel). - For models like Gemini that provide thinking tokens, they now display in **Token Usage** in the Weave UI. Thinking tokens are internal tokens used by LLMs to process complex reasoning steps, calculations, and self-reflection before generating a final answer. ## Fixes @@ -1039,8 +1039,8 @@ In v0.73.0, resizing or reordering workspace panels may not be reflected in save ### Weave - You can now group sessions or conversations across multiple traces with [Weave Threads](/weave/guides/tracking/threads). - The **Trace view**'s new [Graph view](/weave/guides/tracking/trace-tree#graph-view) visualizes the call tree structure as a graph. -- The **Trace view** includes new specialized views for [documents](/weave/guides/core-types/media#documents) retrieved with ChromaDB and Langchain improves handling of schemas such as chats. In **Trace view**, click a document to open its detailed view in a drawer to the right. Learn more at [Navigate the trace view](/weave/guides/tracking/trace-tree). -- You can now attach [HTML](/weave/guides/core-types/media#html) to a trace as a media type. +- The **Trace view** includes new specialized views for [documents](/weave/guides/core-types/media#log-documents) retrieved with ChromaDB and Langchain improves handling of schemas such as chats. In **Trace view**, click a document to open its detailed view in a drawer to the right. Learn more at [Navigate the trace view](/weave/guides/tracking/trace-tree). +- You can now attach [HTML](/weave/guides/core-types/media#log-html) to a trace as a media type. - From the **Assets** tab, you can now create and edit prompts. Click **Assets**, then in the navigation, click **Prompts**. Click **New prompt** or click the name of an existing prompt to view its details. From there, click the pencil icon to edit and republish the prompt. ## Fixes diff --git a/support/models/articles/how-do-i-enable-code-logging-with-sweeps.mdx b/support/models/articles/how-do-i-enable-code-logging-with-sweeps.mdx index 12f5b4a775..1a16768d0b 100644 --- a/support/models/articles/how-do-i-enable-code-logging-with-sweeps.mdx +++ b/support/models/articles/how-do-i-enable-code-logging-with-sweeps.mdx @@ -3,7 +3,7 @@ title: "How do I enable code logging with Sweeps?" keywords: ["Sweeps"] --- -To enable code logging for sweeps, add `wandb.log_code()` after initializing the W&B Run. This step is necessary even when you enable code logging in your W&B profile settings. For advanced code logging, see the [`wandb.log_code()` reference](/models/ref/python/experiments/run#log_code). +To enable code logging for sweeps, add `wandb.log_code()` after initializing the W&B Run. This step is necessary even when you enable code logging in your W&B profile settings. For advanced code logging, see the [`wandb.log_code()` reference](/models/ref/python/experiments/run#method-run-log_code). --- diff --git a/support/models/articles/how-do-i-log-a-list-of-values.mdx b/support/models/articles/how-do-i-log-a-list-of-values.mdx index f2a78afc70..ad87abe1f7 100644 --- a/support/models/articles/how-do-i-log-a-list-of-values.mdx +++ b/support/models/articles/how-do-i-log-a-list-of-values.mdx @@ -3,7 +3,7 @@ title: "How do I log a list of values?" keywords: ["Logs", "Experiments"] --- -You can log a list of values, such as per-step losses, with [`wandb.Run.log()`](/models/ref/python/experiments/run/#method-runlog/). The following examples show two common approaches: log the list as a dictionary entry, or log it as a histogram to visualize its distribution. +You can log a list of values, such as per-step losses, with [`wandb.Run.log()`](/models/ref/python/experiments/run#method-run-log). The following examples show two common approaches: log the list as a dictionary entry, or log it as a histogram to visualize its distribution. diff --git a/support/models/articles/how-do-i-log-metrics-on-two-time-scales.mdx b/support/models/articles/how-do-i-log-metrics-on-two-time-scales.mdx index 97f2ea7286..a1e56383c7 100644 --- a/support/models/articles/how-do-i-log-metrics-on-two-time-scales.mdx +++ b/support/models/articles/how-do-i-log-metrics-on-two-time-scales.mdx @@ -3,7 +3,7 @@ title: "How do I log metrics on two different time scales?" keywords: ["Experiments", "Metrics"] --- -To log metrics on two different time scales, log indices like `batch` and `epoch` alongside your metrics. For example, you might log training accuracy per batch and validation accuracy per epoch. Call `run.log({'train_accuracy': 0.9, 'batch': 200})` in one step and `run.log({'val_accuracy': 0.8, 'epoch': 4})` in another. In the UI, set the value you want as the x-axis for each chart. To set a default x-axis for a specific index, use [`Run.define_metric()`](/models/ref/python/experiments/run#define_metric). For the preceding example, use the following code: +To log metrics on two different time scales, log indices like `batch` and `epoch` alongside your metrics. For example, you might log training accuracy per batch and validation accuracy per epoch. Call `run.log({'train_accuracy': 0.9, 'batch': 200})` in one step and `run.log({'val_accuracy': 0.8, 'epoch': 4})` in another. In the UI, set the value you want as the x-axis for each chart. To set a default x-axis for a specific index, use [`Run.define_metric()`](/models/ref/python/experiments/run#method-run-define_metric). For the preceding example, use the following code: ```python import wandb diff --git a/support/models/articles/how-do-i-move-a-run-to-another-project.mdx b/support/models/articles/how-do-i-move-a-run-to-another-project.mdx index f9937bc38e..3f0a5a7822 100644 --- a/support/models/articles/how-do-i-move-a-run-to-another-project.mdx +++ b/support/models/articles/how-do-i-move-a-run-to-another-project.mdx @@ -13,7 +13,7 @@ You can move a run from one project to another to reorganize your work, consolid The selected runs now appear in the destination project. -W&B supports moving runs through the UI, but doesn't support copying runs. Artifacts logged with the runs don't transfer to the new project. To manually move artifacts to the run's new location, use the [`wandb artifact get`](/models/ref/cli/wandb-artifact/wandb-artifact-get/) SDK command or the [`Api.artifact` API](/models/ref/python/public-api/api/#artifact) to download the artifact, then use [`wandb artifact put`](/models/ref/cli/wandb-artifact/wandb-artifact-put/) or the `Api.artifact` API to upload it to the run's new location. +W&B supports moving runs through the UI, but doesn't support copying runs. Artifacts logged with the runs don't transfer to the new project. To manually move artifacts to the run's new location, use the [`wandb artifact get`](/models/ref/cli/wandb-artifact/wandb-artifact-get/) SDK command or the [`Api.artifact` API](/models/ref/python/public-api/api#method-api-artifact) to download the artifact, then use [`wandb artifact put`](/models/ref/cli/wandb-artifact/wandb-artifact-put/) or the `Api.artifact` API to upload it to the run's new location. --- diff --git a/support/models/articles/workspace-loads-slowly-with-many-metric.mdx b/support/models/articles/workspace-loads-slowly-with-many-metric.mdx index bbff2a6d29..5012afff77 100644 --- a/support/models/articles/workspace-loads-slowly-with-many-metric.mdx +++ b/support/models/articles/workspace-loads-slowly-with-many-metric.mdx @@ -16,7 +16,7 @@ For runs that log between 5,000 and 100,000 metrics, switch to a [manual workspa If an automatic workspace includes panels for many logged keys, it might load slowly. To fix this, reduce the panel count by switching to manual mode and adding back only the panels you need: 1. Reset the workspace to manual mode. -2. Use [Quick add](/models/app/features/panels/#quick-add) to selectively add panels for the metrics you need. +2. Use [Quick panel builder](/models/app/features/panels#quick-panel-builder) to selectively add panels for the metrics you need. After you complete these steps, your workspace displays only the panels you explicitly added, which should improve load times. diff --git a/weave/guides/core-types/evaluations.mdx b/weave/guides/core-types/evaluations.mdx index 0632ed5acf..629a1ba112 100644 --- a/weave/guides/core-types/evaluations.mdx +++ b/weave/guides/core-types/evaluations.mdx @@ -165,7 +165,7 @@ Then, create one or more [scoring functions](../evaluation/scorers). Weave uses In some applications, you may want to create custom `Scorer` classes. For example, a standardized `LLMJudge` class can be created with specific parameters (for example, chat model, prompt), specific scoring of each row, and specific calculation of an aggregate score. - See the tutorial on defining a `Scorer` class in [Model-Based Evaluation of RAG applications](/weave/tutorial-rag#optional-defining-a-scorer-class) for more information. + See the tutorial on defining a `Scorer` class in [Model-Based Evaluation of RAG applications](/weave/tutorial-rag#optional-define-a-scorer-class) for more information. Each scoring function is wrapped with `weave.op` and receives an object with `modelOutput` and `datasetRow` properties. diff --git a/weave/guides/evaluation/scorers.mdx b/weave/guides/evaluation/scorers.mdx index a3ddc83a87..d9e4ba575e 100644 --- a/weave/guides/evaluation/scorers.mdx +++ b/weave/guides/evaluation/scorers.mdx @@ -340,7 +340,7 @@ Return a JSON object with your rating and reasoning. > In this example, the default `auto_summarize` would return the count and proportion of True. - For more information, see the implementation of [CorrectnessLLMJudge](/weave/tutorial-rag#optional-defining-a-scorer-class). + For more information, see the implementation of [CorrectnessLLMJudge](/weave/tutorial-rag#optional-define-a-scorer-class). diff --git a/weave/guides/tools/weave-in-workspaces.mdx b/weave/guides/tools/weave-in-workspaces.mdx index 7b4f05a341..41e9c88952 100644 --- a/weave/guides/tools/weave-in-workspaces.mdx +++ b/weave/guides/tools/weave-in-workspaces.mdx @@ -107,7 +107,7 @@ def my_function(x): my_function(10) # This trace is linked to "my-run-id" at step 5. ``` -The `run_id` is the run's short ID (for example, `38m4t5ja`), not the full `entity/project/run_id` path. You can find a run's short ID in the run URL (`https://wandb.ai/team/project/runs/[RUN-ID]`) or by [accessing `wandb.run.id` in code](/models/runs/run-identifiers#find-and-customize-a-runs-id-or-name). +The `run_id` is the run's short ID (for example, `38m4t5ja`), not the full `entity/project/run_id` path. You can find a run's short ID in the run URL (`https://wandb.ai/team/project/runs/[RUN-ID]`) or by [accessing `wandb.run.id` in code](/models/runs/run-identifiers#find-a-runs-id). To stop the override and fall back to the global `wandb.run`, call `clear_wandb_run_context`: diff --git a/weave/tutorial-eval.mdx b/weave/tutorial-eval.mdx index 26359b5bdf..2dcba9e332 100644 --- a/weave/tutorial-eval.mdx +++ b/weave/tutorial-eval.mdx @@ -257,7 +257,7 @@ When you use Weave evaluations, Weave expects a `target` to compare `output` aga For more information on making your own scoring function, see the [Scorers](/weave/guides/evaluation/scorers) guide. -In some applications, you may want to create custom `Scorer` classes. For example, you might create a standardized `LLMJudge` class with specific parameters (such as chat model or prompt), specific row scoring, and aggregate score calculation. For more information, see the tutorial on defining a `Scorer` class in [Model-based evaluation of RAG applications](/weave/tutorial-rag#optional-defining-a-scorer-class). +In some applications, you may want to create custom `Scorer` classes. For example, you might create a standardized `LLMJudge` class with specific parameters (such as chat model or prompt), specific row scoring, and aggregate score calculation. For more information, see the tutorial on defining a `Scorer` class in [Model-based evaluation of RAG applications](/weave/tutorial-rag#optional-define-a-scorer-class). ## Use a built-in scorer and run the evaluation