From 74c6ab5bcfa13842602edbb340da814ab5106ba3 Mon Sep 17 00:00:00 2001 From: John Mulhausen Date: Thu, 4 Jun 2026 14:49:56 -0400 Subject: [PATCH] style: apply /style-guide pass to snippets/_includes --- .../_includes/api-key-create-streamlined.mdx | 2 +- snippets/_includes/api-key-create.mdx | 2 +- snippets/_includes/api-key-find.mdx | 4 +- snippets/_includes/api-key-security.mdx | 42 +++++++------ .../_includes/api-key-view-once-warning.mdx | 2 +- snippets/_includes/byob-context-note.mdx | 2 +- snippets/_includes/byob-provisioning-link.mdx | 14 ++--- snippets/_includes/cli/README.md | 43 +++++++------ snippets/_includes/cli/wandb-beta-leet.mdx | 62 ++++++++++--------- .../runs_update_tag_public_api.mdx | 19 +++--- .../_includes/line-plot-capture-groups.mdx | 17 ++--- .../llm-eval-jobs/export-evaluation.mdx | 5 +- snippets/_includes/llm-eval-jobs/preview.mdx | 2 +- .../llm-eval-jobs/rerun-evaluation.mdx | 13 ++-- .../review-evaluation-results.mdx | 42 +++++++------ .../_includes/multi-tenant-cloud-only.mdx | 2 +- .../object-storage-configuration-intro.mdx | 8 +-- .../_includes/org-service-account-create.mdx | 4 +- .../pin-runs-condensed.mdx | 6 +- .../_includes/private-preview-feature.mdx | 2 +- .../_includes/project-visibility-settings.mdx | 10 +-- snippets/_includes/public-api-use.mdx | 2 +- .../rate-limits-defaults-and-notification.mdx | 7 ++- .../release-notes-support-eol-reminder.mdx | 2 +- .../self-managed-hardware-requirements.mdx | 4 +- .../self-managed-mysql-requirements.mdx | 6 +- .../self-managed-networking-requirements.mdx | 15 +++-- ...lf-managed-object-storage-requirements.mdx | 15 ++--- .../_includes/self-managed-obtain-license.mdx | 11 ++-- .../self-managed-redis-requirements.mdx | 4 +- .../self-managed-ssl-tls-requirements.mdx | 8 +-- .../self-managed-verify-installation.mdx | 44 +++++++------ ...pi-key-create-additional-single-tenant.mdx | 4 +- .../service-account-api-key-delete.mdx | 6 +- .../_includes/service-account-benefits.mdx | 12 ++-- .../_includes/team-service-account-create.mdx | 4 +- .../_includes/weave-quickstart-prereqs.mdx | 8 ++- 37 files changed, 244 insertions(+), 211 deletions(-) diff --git a/snippets/_includes/api-key-create-streamlined.mdx b/snippets/_includes/api-key-create-streamlined.mdx index aab568dbd4..cfca192116 100644 --- a/snippets/_includes/api-key-create-streamlined.mdx +++ b/snippets/_includes/api-key-create-streamlined.mdx @@ -1,3 +1,3 @@ -For a more streamlined approach, create an API key by going directly to [User Settings](https://wandb.ai/settings). Copy the newly created API key immediately and save it in a secure location such as a password manager. +For a more streamlined approach, create an API key directly in [User Settings](https://wandb.ai/settings). Copy the new API key immediately and save it in a secure location such as a password manager. diff --git a/snippets/_includes/api-key-create.mdx b/snippets/_includes/api-key-create.mdx index b3330b0a49..b07afbc919 100644 --- a/snippets/_includes/api-key-create.mdx +++ b/snippets/_includes/api-key-create.mdx @@ -1,7 +1,7 @@ import ApiKeyViewOnceWarning from "/snippets/_includes/api-key-view-once-warning.mdx"; import ServiceAccountApiKeyCreate from "/snippets/_includes/service-account-api-key-create-additional.mdx"; -To create an API key, select the **Personal API key** or **Service Account API key** tab for details. +To create an API key, select the **Personal API key** or **Service account API key** tab. diff --git a/snippets/_includes/api-key-find.mdx b/snippets/_includes/api-key-find.mdx index 9b1183e791..8d9e41c679 100644 --- a/snippets/_includes/api-key-find.mdx +++ b/snippets/_includes/api-key-find.mdx @@ -18,10 +18,10 @@ To find a personal API key owned by your user ID: To find an API key owned by an organization or team service account: -1. Navigate to the **API Keys** tab in your organization settings. +1. In your organization settings, navigate to the **API Keys** tab. 2. Find the API key in the list. You can search or filter the list by owner, key name, or key ID. -The API keys table shows the key ID (the first part of each API key) for identification purposes. The full secret API key is only displayed once when you create it. If you need to use an existing key but do not have the full secret stored, you must create a new API key. +The API keys table shows the key ID (the first part of each API key) for identification. The full secret API key is only displayed once when you create it. If you need to use an existing key but don't have the full secret stored, you must create a new API key. diff --git a/snippets/_includes/api-key-security.mdx b/snippets/_includes/api-key-security.mdx index 7ae4097c9d..2c4c554db5 100644 --- a/snippets/_includes/api-key-security.mdx +++ b/snippets/_includes/api-key-security.mdx @@ -1,54 +1,58 @@ ## Store and handle API keys securely -API keys provide access to your W&B account and should be protected like passwords. Follow these best practices: +API keys provide access to your W&B account and you must protect them like passwords. The following sections describe recommended storage methods, practices to avoid, and how to handle API keys in your code so that you can reduce the risk of accidental exposure. ### Recommended storage methods +Choose a storage method based on whether you use the key in production systems or on a local workstation. We recommend the following options: + - **Secrets manager**: Use a dedicated secrets management system such as [AWS Secrets Manager](https://aws.amazon.com/secrets-manager/), [HashiCorp Vault](https://developer.hashicorp.com/vault), [Azure Key Vault](https://azure.microsoft.com/en-us/products/key-vault), or [Google Secret Manager](https://cloud.google.com/security/products/secret-manager). - **Password manager**: Use a reputable password manager application. -- **OS-level keychains**: Store keys in macOS Keychain, Windows Credential Manager, or Linux secret service. Not suggested for production. +- **OS-level keychains**: Store keys in macOS Keychain, Windows Credential Manager, or Linux secret service. We don't recommend this option for production. ### What to avoid -- Never commit API keys to version control systems such as Git. -- Do not store API keys in plain text configuration files. -- Do not pass API keys on the command line, because they will be visible in the output of OS commands like `ps`. -- Avoid sharing API keys through email, chat, or other unencrypted channels. -- Do not hard-code API keys in your source code. +The following practices can expose API keys to other users, processes, or public systems. You must not: + +- Commit API keys to version control systems such as Git. +- Store API keys in plain text configuration files. +- Pass API keys on the command line, because they're visible in the output of OS commands like `ps`. +- Share API keys through email, chat, or other unencrypted channels. +- Hard-code API keys in your source code. -If an API key is exposed, delete the API key from your W&B account immediately and contact [support](mailto:support@wandb.ai) or your AISE. +If an API key is exposed, delete it from your W&B account immediately and contact [support](mailto:support@wandb.ai) or your AISE. ### Environment variables -When using API keys in your code, pass them through environment variables: +When you use API keys in your code, pass them through environment variables so that the key stays out of source files and is easier to rotate. Replace `[API-KEY]` with your W&B API key: ```bash -export WANDB_API_KEY="your-api-key-here" +export WANDB_API_KEY="[API-KEY]" ``` -This approach keeps keys out of your source code and makes it easier to rotate them when needed. - -Avoid setting the environment variable in line with the command, because it will be visible in the output of OS commands like `ps`: +Avoid setting the environment variable on the same line as the command, because it's visible in the output of OS commands like `ps`. Replace `[API-KEY]` with your W&B API key: ```bash # Avoid this pattern, which can expose the API key in process managers -export WANDB_API_KEY="your-api-key-here" ./my-script.sh +export WANDB_API_KEY="[API-KEY]" ./my-script.sh ``` ### SDK version compatibility -New API keys are longer than legacy keys. When authenticating with older versions of the `wandb` or `weave` SDKs, you may encounter an API key length error. +New API keys are longer than legacy keys. When you authenticate with older versions of the `wandb` or `weave` SDKs, you might encounter an API key length error. To resolve this error, update to a newer SDK version: -**Solution**: Update to a newer SDK version: -- `wandb` SDK v0.22.3+ +- `wandb` SDK v0.22.3+: ```bash pip install --upgrade wandb==0.22.3 ``` -- `weave` SDK v0.52.17+ + +- `weave` SDK v0.52.17+: + ```bash pip install --upgrade weave==0.52.17 ``` -If you cannot upgrade the SDK immediately, set the API key using the `WANDB_API_KEY` environment variable as a workaround. + +If you can't upgrade the SDK immediately, set the API key using the `WANDB_API_KEY` environment variable as a workaround. diff --git a/snippets/_includes/api-key-view-once-warning.mdx b/snippets/_includes/api-key-view-once-warning.mdx index df753a08df..195ac5d592 100644 --- a/snippets/_includes/api-key-view-once-warning.mdx +++ b/snippets/_includes/api-key-view-once-warning.mdx @@ -1,3 +1,3 @@ -The full API key is only shown once at creation time. After you close the dialog, you cannot view the full API key again. Only the key ID (first part of the key) is visible in your settings. If you lose the full API key, you must create a new API key. +The full API key appears only once, when you create it. After you close the dialog, you can't view the full API key again. Your settings show only the key ID (the first part of the key). If you lose the full API key, create a new one. diff --git a/snippets/_includes/byob-context-note.mdx b/snippets/_includes/byob-context-note.mdx index 0cae9fedb1..360381ad59 100644 --- a/snippets/_includes/byob-context-note.mdx +++ b/snippets/_includes/byob-context-note.mdx @@ -1,5 +1,5 @@ -**This guide applies to all W&B deployment types:** +This guide applies to all W&B deployment types: - **Multi-tenant Cloud**: Team-level BYOB - **Dedicated Cloud**: Instance and team-level BYOB - **Self-Managed**: Instance and team-level BYOB diff --git a/snippets/_includes/byob-provisioning-link.mdx b/snippets/_includes/byob-provisioning-link.mdx index b85b92bfdf..51d9fc6825 100644 --- a/snippets/_includes/byob-provisioning-link.mdx +++ b/snippets/_includes/byob-provisioning-link.mdx @@ -1,10 +1,10 @@ ### Provision your storage bucket -Before configuring W&B, provision your object storage bucket with proper IAM policies, CORS configuration, and access credentials. +Before you configure W&B, you must provision your object storage bucket with the appropriate IAM policies, CORS configuration, and access credentials. This ensures that W&B can read from and write to your bucket after you configure the connection. -**See the [Bring Your Own Bucket (BYOB) guide](/platform/hosting/data-security/secure-storage-connector) for detailed step-by-step provisioning instructions for:** -- Amazon S3 (including IAM policies and bucket policies) -- Google Cloud Storage (including PubSub notifications) -- Azure Blob Storage (including managed identities) -- CoreWeave AI Object Storage -- S3-compatible storage (MinIO Enterprise, NetApp StorageGRID, and other enterprise solutions) +See the [Bring Your Own Bucket (BYOB) guide](/platform/hosting/data-security/secure-storage-connector) for step-by-step provisioning instructions for the following storage providers: +- Amazon S3 (including IAM policies and bucket policies). +- Google Cloud Storage (including Pub/Sub notifications). +- Azure Blob Storage (including managed identities). +- CoreWeave AI Object Storage. +- S3-compatible storage (MinIO Enterprise, NetApp StorageGRID). diff --git a/snippets/_includes/cli/README.md b/snippets/_includes/cli/README.md index b60793d78c..1d24882012 100644 --- a/snippets/_includes/cli/README.md +++ b/snippets/_includes/cli/README.md @@ -1,21 +1,23 @@ -# CLI Reference Intro Snippets +# CLI reference intro snippets -This directory contains introductory content for W&B CLI commands that gets included in the auto-generated CLI reference pages. +This directory contains introductory content for W&B CLI commands that's included in the auto-generated CLI reference pages. This README is for docs contributors who want to add or update intro content for a CLI command reference page. ## Purpose -The CLI reference pages (`models/ref/cli/`) are auto-generated from the W&B CLI source code using `scripts/cli-docs-generator.py`. However, some commands benefit from additional context, examples, or explanations beyond what's in the CLI help text. This directory holds that supplementary content. +The `scripts/cli-docs-generator.py` script auto-generates the CLI reference pages in `models/ref/cli/` from the W&B CLI source code. However, some commands benefit from additional context, examples, or explanations beyond what's in the CLI help text. This directory holds that supplementary content. -## How It Works +## How it works -1. **Snippet files** are created here with the naming pattern: `wandb-{command}.mdx` +The generation process works as follows: + +1. Create **snippet files** here with the naming pattern `wandb-[COMMAND].mdx`. - Example: `wandb-login.mdx`, `wandb-sync.mdx`, `wandb-verify.mdx` - For nested commands, use the full command with dashes: `wandb-artifact-get.mdx` -2. **Generation script** automatically detects these snippets and: - - Adds an import statement at the top of the generated page - - Includes the snippet content right after the front matter, before the Usage section - - For pages without snippets, adds a commented-out template showing how to add one +2. The **generation script** automatically detects these snippets and: + - Adds an import statement at the top of the generated page. + - Includes the snippet content immediately after the front matter, before the Usage section. + - For pages without snippets, adds a commented-out template that shows how to add one. 3. **Generated page structure**: ```markdown @@ -30,19 +32,22 @@ The CLI reference pages (`models/ref/cli/`) are auto-generated from the W&B CLI [auto-generated content from CLI] ``` -## Adding Intro Content for a Command +## Add intro content for a command -To add introductory content for a command: +Follow these steps to add introductory content that appears above the auto-generated reference for a CLI command. -1. Create a new snippet with the correct naming: `wandb-{command-name}.mdx`. -2. Write the intro content in MDX format (can include markdown, code blocks, etc.). -3. In the generated CLI reference for the command, uncomment the lines to include and use the snippet. No need to regenerate all of the references. -4. Next time the references are generated, the generator will automatically detect and include your snippet. +1. Create a new snippet with the correct naming, `wandb-[COMMAND-NAME].mdx`. The generator uses this filename to match the snippet to its command. +2. Write the intro content in MDX format, which can include Markdown and code blocks. +3. In the generated CLI reference for the command, uncomment the lines to include and use the snippet. You don't need to regenerate all of the references. +4. The next time you generate the references, the generator automatically detects and includes your snippet. +Once you complete these steps, the snippet appears on the command's reference page immediately above the Usage section, and persists across subsequent regenerations. ## Guidelines -- Keep content concise and focused on what users need to know before the command details -- Use code blocks for examples -- Avoid duplicating information that's already in the auto-generated command options/arguments -- Test locally after adding a snippet to ensure it renders correctly +Apply the following guidelines when authoring snippet content to keep intro sections consistent across CLI reference pages. + +- Keep content concise and focused on what users need to know before the command details. +- Use code blocks for examples. +- Snippets must avoid duplicating information that's already in the auto-generated command options or arguments. +- After you add a snippet, test locally to ensure it renders correctly. diff --git a/snippets/_includes/cli/wandb-beta-leet.mdx b/snippets/_includes/cli/wandb-beta-leet.mdx index 52decf4cda..26ff478b86 100644 --- a/snippets/_includes/cli/wandb-beta-leet.mdx +++ b/snippets/_includes/cli/wandb-beta-leet.mdx @@ -1,6 +1,6 @@ W&B LEET, the Lightweight Experiment Exploration Tool, is a local terminal UI for exploring W&B runs from the command line. Use it to compare runs, inspect metric charts, browse logged images, tail console logs, and watch system metrics without leaving a terminal session. -LEET reads local `.wandb` run files, so you can inspect runs before syncing them to W&B. LEET is a keyboard-driven interface designed for users who work on remote systems or HPC environments using tools like SSH or `tmux`. +LEET reads local `.wandb` run files, so you can inspect runs before syncing them to W&B. LEET is a keyboard-driven interface for users who work on remote systems or HPC environments with tools like SSH or `tmux`. @@ -9,6 +9,8 @@ LEET reads local `.wandb` run files, so you can inspect runs before syncing them ## Examples +The following examples show common ways to launch LEET. + Open the default workspace: ```bash @@ -49,9 +51,9 @@ wandb beta leet config LEET has three main views: workspace, single-run, and SYMON. -### Workspace View +### Workspace view -Workspace view is the default LEET experience. It is built for comparing multiple runs from the same local `wandb/` directory. +Workspace view is the default LEET view. It's built for comparing multiple runs from the same local `wandb/` directory. Workspace view includes: @@ -62,18 +64,18 @@ Workspace view includes: - A console logs pane. - A run overview sidebar with state, ID, name, project, tags, notes, environment, config, and summary values. -Use `space` to select or deselect a run. Selecting a run adds it to the overlaid metric charts and loads its run-specific data. Use `p` to pin or unpin a run. Pinning keeps that run's metric series on top in overlaid metric charts; the run overview, system metrics, media, and console logs panes follow the run highlighted in the runs sidebar. Press `enter` on the highlighted run to open it in single-run view. +Use `space` to select or deselect a run. When you select a run, LEET adds it to the overlaid metric charts and loads its run-specific data. Use `p` to pin or unpin a run. Pinning keeps that run's metric series on top in overlaid metric charts. The run overview, system metrics, media, and console logs panes follow the run highlighted in the runs sidebar. Press `enter` on the highlighted run to open it in single-run view. Selected live runs continue updating in the workspace, so you can use LEET for both post-run analysis and live monitoring. -### Single-Run View +### Single-run view -Single-run view focuses on one run. It uses the same chart engine as workspace view, but is arranged around that run: +Single-run view focuses on one run. It uses the same chart engine as workspace view, but arranges the panes around that run: - The main metrics grid is centered. - The run overview sidebar is on the left. - The system metrics sidebar is on the right. -- Media and console logs panes can be opened below the metrics grid. +- You can open media and console logs panes below the metrics grid. W&B LEET single-run view with run overview, metric charts, and system metrics panes. @@ -83,9 +85,9 @@ Press `esc` to return from single-run view to the workspace. ### SYMON -`wandb beta leet symon` opens a standalone system monitor that is not tied to a W&B run. +`wandb beta leet symon` opens a standalone system monitor that isn't tied to a W&B run. -Use SYMON to watch local CPU, memory, disk, network, GPU, TPU, IPU, and Trainium metrics while another process is running. It uses the same system metrics chart engine as the workspace and single-run views. +Use SYMON to watch local CPU, memory, disk, network, GPU, TPU, IPU, and Trainium metrics while another process runs. It uses the same system metrics chart engine as the workspace and single-run views. Set the sampling cadence with `--interval`: @@ -99,15 +101,17 @@ wandb beta leet symon --interval 1m W&B LEET system metrics view with GPU charts and bucketed heatmap mode. -## Key Features +## Key features + +The following sections describe the panes, filters, and chart modes that make up the LEET interface. -### Multi-Run Comparison +### Multi-run comparison -Workspace mode lets you select multiple local runs and compare their metric series on the same charts. Each selected run gets a stable color. LEET also avoids base color collisions so visible runs are easier to distinguish. +Workspace view lets you select multiple local runs and compare their metric series on the same charts. LEET assigns each selected run a stable color and avoids base color collisions so visible runs are easier to distinguish. The runs list uses distinct markers for unselected, selected, and pinned runs. Pinning is useful when many runs are selected and you want one run's metric series to render on top. -### Run Filtering +### Run filtering Press `f` in workspace view to filter the runs sidebar. Bare terms search across run key, display name, run ID, project, tags, and notes. @@ -141,7 +145,7 @@ project:vision tag:baseline cfg.lr>=1e-3 -note:debug | project:nlp While editing a filter, press `tab` to switch between regex and glob modes. Regex mode behaves like a case-insensitive substring search unless the query contains regex metacharacters. Glob mode supports `*` for any sequence and `?` for any single character. -### Metrics and System Metrics +### Metrics and system metrics LEET renders scalar metrics as terminal line charts. Use `/` to filter run metrics and `\` to filter system metrics. Press `ctrl+/` to clear the metrics filter and `ctrl+\` to clear the system metrics filter. @@ -153,13 +157,13 @@ Press `y` on a focused chart to cycle chart modes: - System metric charts toggle log-scale Y. - Percentage-based system metric charts can also switch into bucketed heatmap mode. -Use the mouse wheel to zoom the focused chart. On live system metric charts, LEET defaults to a rolling tail window, which is 10 minutes by default. Zooming out reveals more history. +Use the mouse wheel to zoom the focused chart. On live system metric charts, LEET defaults to a rolling 10-minute tail window. Zoom out to reveal more history. Right-click and drag on a chart to inspect the nearest point. Hold `alt` while right-clicking and dragging to inspect all visible charts at the same X position. -### Media Pane +### Media pane -The media pane renders `wandb.Image` data as ANSI thumbnails directly in the terminal. It is available in both workspace and single-run views. +The media pane renders `wandb.Image` data as ANSI thumbnails directly in the terminal. It's available in both workspace and single-run views. Open or close the media pane with `3`. When the media pane is focused: @@ -176,13 +180,13 @@ Open or close the media pane with `3`. When the media pane is focused: W&B LEET workspace view with the media pane showing several image series. -### Console Logs +### Console logs Open or close the console logs pane with `4`. The pane shows console output for the open run in single-run view, or for the highlighted run in workspace view. LEET assembles raw terminal output into readable log lines, including output that contains ANSI escape codes, partial lines, and carriage returns. -### Run Overview +### Run overview The run overview sidebar shows run metadata and logged values: @@ -193,7 +197,7 @@ The run overview sidebar shows run metadata and logged values: Press `o` to filter overview items. Press `ctrl+o` to clear the overview filter. -### System Metrics Coverage +### System metrics coverage The system metrics pane and SYMON can display host and accelerator metrics when those metrics are available from the local run file or live sampler. Supported chart definitions include: @@ -211,13 +215,13 @@ The system metrics pane and SYMON can display host and accelerator metrics when ## Configuration -Run the config editor with: +LEET reads persistent settings such as startup mode, grid sizes, default pane visibility, and color schemes from a config file. Run the config editor with: ```bash wandb beta leet config ``` -LEET stores configuration in `wandb-leet.json`. By default, the file is written under: +LEET stores configuration in `wandb-leet.json`. By default, LEET writes the file under: ```text ~/.config/wandb/wandb-leet.json @@ -248,11 +252,11 @@ You can also resize the focused grid from inside LEET: After pressing `c` or `r`, press a digit from `1` to `9`, or press `esc` to cancel. -### Common Config Keys +### Common config keys | Key | Default | Description | | --- | --- | --- | -| `startup_mode` | `workspace_latest` | Initial view when launched without a run path. | +| `startup_mode` | `workspace_latest` | Initial view when you launch LEET without a run path. | | `metrics_grid.rows`, `metrics_grid.cols` | `4`, `3` | Single-run metrics grid size. | | `system_grid.rows`, `system_grid.cols` | `6`, `2` | Single-run system metrics sidebar grid size. | | `media_grid.rows`, `media_grid.cols` | `1`, `2` | Single-run media grid size. | @@ -284,9 +288,9 @@ Available color schemes include `wandb-vibe-10`, `wandb-vibe-20`, `sunset-glow`, `dusk-shore` and `clear-signal` are colorblind-friendly palettes. Sequential palettes such as `viridis`, `plasma`, `inferno`, `magma`, `cividis`, and `traffic-light` work well for bucketed heatmaps. -## Keyboard Shortcuts +## Keyboard shortcuts -Press `h` or `?` inside LEET to open the in-app help screen. Shortcuts vary slightly by view. +The following tables list the keyboard and mouse shortcuts available in each view. Press `h` or `?` inside LEET to open the in-app help screen. Shortcuts vary slightly by view. ### Workspace @@ -323,7 +327,7 @@ Press `h` or `?` inside LEET to open the in-app help screen. Shortcuts vary slig | `home` | Jump to first item, first page, or first media frame. | | `end` | Jump to last item, last page, or latest media frame. | -### Single-Run +### Single-run | Key | Action | | --- | --- | @@ -382,7 +386,7 @@ Press `h` or `?` inside LEET to open the in-app help screen. Shortcuts vary slig ## Changelog -The following table summarizes LEET changes by version. For more details, see the [GitHub release notes](https://github.com/wandb/client/releases). +The following table summarizes LEET changes by version. For more information, see the [GitHub release notes](https://github.com/wandb/client/releases). | Version | LEET changes | | --- | --- | @@ -390,5 +394,5 @@ The following table summarizes LEET changes by version. For more details, see th | `0.23.1` | Regex filters for metrics and the run overview, chart inspection with right-click drag, synchronized chart inspection with `alt`, and better negative Y-axis tick display and system metric units. | | `0.25.0` | Multi-run workspace view and `wandb beta leet config` editor. | | `0.25.1` | Console logs pane, workspace system metrics pane, and system metrics filtering. | -| `0.26.0` | Metadata-aware run filtering, tags and notes in overview, per-chart log Y, SYMON, bucketed heatmaps, colorblind-friendly palettes, media pane, improved system metrics overlays/inspection/zoom, and workspace run color collision prevention. | +| `0.26.0` | Metadata-aware run filtering, tags and notes in overview, per-chart log Y, SYMON, bucketed heatmaps, colorblind-friendly palettes, media pane, improved system metrics (overlays, inspection, and zoom), and workspace run color collision prevention. | | `0.26.1` | Unified navigation across focused panes with `w`/`a`/`s`/`d`, arrows, `home`, `end`, `pgup`, and `pgdown`. | diff --git a/snippets/_includes/code-examples/runs_update_tag_public_api.mdx b/snippets/_includes/code-examples/runs_update_tag_public_api.mdx index 8176eff714..31709d5884 100644 --- a/snippets/_includes/code-examples/runs_update_tag_public_api.mdx +++ b/snippets/_includes/code-examples/runs_update_tag_public_api.mdx @@ -2,22 +2,21 @@ """ Add one or more tags to previously saved runs. -Use the Public API to update tags on stored data after the run has finished or -when you are working outside the run process. +Use the Public API to update tags after the run finishes or when you're +working outside the run process. -To add a new tag to a run, update the "tags" -property with a new list that includes the existing tags and the new tag(s). -The run's path consists of entity/project/run_id -After updating the tags, call the run's `update()` method to save the changes. +To add a new tag to a run, append it to the run's `tags` list, then call the +run's `update()` method to save the changes. The run's path consists of +`entity/project/run_id`. """ import wandb -entity = "" -project = "" -run_id = "" # Replace with the ID of the run you want to update +entity = "[ENTITY]" +project = "[PROJECT]" +run_id = "[RUN-ID]" # Replace with the ID of the run you want to update # Path consists of entity/project/run_id with wandb.Api().run(f"{entity}/{project}/{run_id}") as run: - run.tags.append("") + run.tags.append("[TAG]") run.update() ``` diff --git a/snippets/_includes/line-plot-capture-groups.mdx b/snippets/_includes/line-plot-capture-groups.mdx index 5e6c43b862..5fda2081b6 100644 --- a/snippets/_includes/line-plot-capture-groups.mdx +++ b/snippets/_includes/line-plot-capture-groups.mdx @@ -1,17 +1,18 @@ -Capture groups in your regular expression control how multi-metric panels are created. This behavior can be confusing if you're not expecting it. +Capture groups in your regular expression control how the UI creates multi-metric panels. This behavior can be confusing if you're not expecting it. - **Capture groups create multiple panels** When your regular expression includes parentheses that form a capture group, the UI creates a separate panel for each unique value captured by that group. - For example, the expression `(layer_0|layer_10)_loss` includes a capture group and will create two separate panels: - 1. One panel for metrics matching `layer_0`. - 2. One panel for metrics matching `layer_10`. + For example, the expression `(layer_0|layer_10)_loss` includes a capture group and creates two separate panels: + - One panel for metrics matching `layer_0`. + - One panel for metrics matching `layer_10`. - **Non-capturing groups keep metrics together** To match multiple alternatives without creating separate panels, use a non-capturing group with the `?:` syntax. The expression `(?:layer_0|layer_10)_loss` matches the same metrics as the previous example, but displays them together in a single panel. -Here's the difference: -- `(layer_0|layer_10)_loss` - Creates two panels, one for each layer. -- `(?:layer_0|layer_10)_loss` - Creates one panel showing both layers together. +The following examples illustrate the difference: -This gives you flexibility to choose the approach that best fits your analysis needs. Use capture groups when you want to separate metrics into distinct panels. Use non-capturing groups when you want to compare metrics together on a single plot. +- `(layer_0|layer_10)_loss`: Creates two panels, one for each layer. +- `(?:layer_0|layer_10)_loss`: Creates one panel showing both layers together. + +Choose the approach that fits your analysis needs. Use capture groups when you want to separate metrics into distinct panels. Use non-capturing groups when you want to compare metrics on a single plot. diff --git a/snippets/_includes/llm-eval-jobs/export-evaluation.mdx b/snippets/_includes/llm-eval-jobs/export-evaluation.mdx index 9df433996c..dc6b97e4b9 100644 --- a/snippets/_includes/llm-eval-jobs/export-evaluation.mdx +++ b/snippets/_includes/llm-eval-jobs/export-evaluation.mdx @@ -1,6 +1,7 @@ ## Export an evaluation job configuration -Export an evaluation job's configuration from the run's **Files** tab. + +To export an evaluation job's configuration from the run's **Files** tab: 1. Open the run in single-run view. 1. Click the **Files** tab. -1. Click the download button next to `config.yaml` to download it locally. +1. Click the download button next to `config.yaml`. diff --git a/snippets/_includes/llm-eval-jobs/preview.mdx b/snippets/_includes/llm-eval-jobs/preview.mdx index 9a01770773..7a548d7281 100644 --- a/snippets/_includes/llm-eval-jobs/preview.mdx +++ b/snippets/_includes/llm-eval-jobs/preview.mdx @@ -1,3 +1,3 @@ -LLM Evaluation Jobs is in **Preview** for [W&B Multi-tenant Cloud](/platform/hosting/hosting-options/multi_tenant_cloud). Compute is free during the preview period. [Learn more](/models/launch#pricing) +LLM Evaluation Jobs is in **Preview** for [W&B Multi-tenant Cloud](/platform/hosting/hosting-options/multi_tenant_cloud). Compute is free during the preview period. [Learn more about Launch pricing](/models/launch#pricing). diff --git a/snippets/_includes/llm-eval-jobs/rerun-evaluation.mdx b/snippets/_includes/llm-eval-jobs/rerun-evaluation.mdx index fbed054297..1ade1a59cb 100644 --- a/snippets/_includes/llm-eval-jobs/rerun-evaluation.mdx +++ b/snippets/_includes/llm-eval-jobs/rerun-evaluation.mdx @@ -1,10 +1,11 @@ -## Re-run an evaluation job -Depending on your situation, there are multiple ways to re-run an evaluation job or view its configuration. +## Rerun an evaluation job -- To re-run the last evaluation job again, follow the steps in [Evaluate your model](#evaluate-your-model). Select the destination project, then the model artifact details and the selected benchmarks you selected last time are populated automatically. Optionally, make adjustments, then launch the evaluation job. -- To re-run an evaluation job from the project's **Runs** tab or run selector, hover over the run name and click the play icon. The job configuration drawer displays with the settings pre-populated. Optionally adjust the settings, then click **Launch**. -- To re-run an evaluation job from a different project, import its configuration: +You can rerun an evaluation job or view its configuration in several ways: + +- To rerun the last evaluation job, follow the steps in [Evaluate your model](#evaluate-your-model). Select the destination project, then the model artifact details and the benchmarks you selected last time populate automatically. Optionally, make adjustments, then launch the evaluation job. +- To rerun an evaluation job from the project's **Runs** tab or run selector, hover over the run name and click the play icon. The job configuration drawer opens with the settings prepopulated. Optionally adjust the settings, then click **Launch**. +- To rerun an evaluation job from a different project, import its configuration: 1. Follow the steps in [Evaluate your model](#evaluate-your-model). After you select the destination project, click **Import configuration**. - 1. Select the project that contains the evaluation job to import, then select the evaluation job run. The job configuration drawer displays with the settings pre-populated. + 1. Select the project that contains the evaluation job to import, then select the evaluation job run. The job configuration drawer opens with the settings prepopulated. 1. Optionally adjust the configuration. 1. Click **Launch**. diff --git a/snippets/_includes/llm-eval-jobs/review-evaluation-results.mdx b/snippets/_includes/llm-eval-jobs/review-evaluation-results.mdx index c967670570..ad511532ce 100644 --- a/snippets/_includes/llm-eval-jobs/review-evaluation-results.mdx +++ b/snippets/_includes/llm-eval-jobs/review-evaluation-results.mdx @@ -1,30 +1,36 @@ ## Review evaluation results -Review your evaluation job results in W&B Models in the destination project's workspace. -1. Click the circular arrow icon at the top of the page to open the recent run modal, where evaluation jobs appear with other runs in the project. If the evaluation job has a leaderboard, click **Leaderboard** to open the leaderboard in full screen, or click a run name to open it in the project in single-run view. +Review your evaluation job results in W&B Models in the destination project's workspace. The following steps describe how to access traces, configuration details, logs, and output files for a completed evaluation job. + +1. Select the circular arrow icon at the top of the page to open the recent run modal, where evaluation jobs appear with other runs in the project. If the evaluation job has a leaderboard, select **Leaderboard** to open the leaderboard in full screen, or select a run name to open it in the project in single-run view. 1. View the evaluation job's traces in the **Evaluations** section of a workspace or in the **Traces** tab of the **Weave** sidebar panel. -1. Click the **Overview** tab to view detailed information about the evaluation job, including its configuration and summary metrics. -1. Click the **Logs** tab to view, search, or download the evaluation job's debug logs. -1. Click the **Files** tab to browse, view, or download the evaluation job's files, including code, log, configuration, and other output files. +1. Select the **Overview** tab to view detailed information about the evaluation job, including its configuration and summary metrics. +1. Select the **Logs** tab to view, search, or download the evaluation job's debug logs. +1. Select the **Files** tab to browse, view, or download the evaluation job's files, including code, log, configuration, and other output files. ## Customize a leaderboard + The leaderboard shows results for all evaluation jobs sent to a given project, with one row per benchmark per evaluation job. Columns display details like the trace, input values, and output values for the evaluation job. For more information about leaderboards, see [Leaderboards in Weave](/weave/guides/core-types/leaderboards). -To give feedback about a result from the leaderboard, click the emoji icon or the chat icon in the **Feedback** column. +To give feedback about a result from the leaderboard, select the emoji icon or the chat icon in the **Feedback** column. + +Customize the leaderboard using the following options: -- By default, all evaluation jobs are displayed. Filter or search for an evaluation job using the run selector at the left. -- By default, evaluation jobs are ungrouped. To group by one or more columns, click the **Group** icon. You can show or hide a group, or expand a group to view its runs. -- By default, all operations are displayed. To display only a single operation, click **All ops** and select an operation. -- To sort by a column, click the column heading. To customize the display of columns, click **Columns**. +- By default, the leaderboard displays all evaluation jobs. Filter or search for an evaluation job using the run selector at the left. +- By default, evaluation jobs are ungrouped. To group by one or more columns, select the **Group** icon. You can show or hide a group, or expand a group to view its runs. +- By default, the leaderboard displays all operations. To display only a single operation, select **All ops** and select an operation. +- To sort by a column, select the column heading. To customize the display of columns, select **Columns**. - By default, headers are organized in a single level. You can increase the header depth to organize related headers together. - - Select or deselect individual columns to show or hide them, or show or hide all columns with a click. + - Select or deselect individual columns to show or hide them, or show or hide all columns at once. - Pin columns to display them before unpinned columns. ## Export a leaderboard -To export a leaderboard: -1. Click the download icon, located near the **Columns** button. -1. To optimize the export size, only the trace roots are exported by default. To export full traces, turn off **Trace roots only**. -1. To optimize the export size, feedback and costs are not exported by default. To include them in the export, toggle **Feedback** or **Costs**. -1. By default, the export is in JSONL format. To customize the format, click **Export to file** and select a format. -1. To export the leaderboard in your browser, click **Export**. -1. To export the leaderboard programmatically, select **Python** or **cURL**, then click **Copy** and run the script or command. + +To export a leaderboard so you can share or further analyze its results outside of W&B, follow these steps: + +1. Select the download icon, located near the **Columns** button. +1. Optional: To optimize the export size, W&B exports only the trace roots by default. To export full traces, turn off **Trace roots only**. +1. Optional: To optimize the export size, W&B excludes feedback and costs from the export by default. To include them in the export, toggle **Feedback** or **Costs**. +1. Optional: By default, W&B exports in JSONL format. To customize the format, select **Export to file** and select a format. +1. To export the leaderboard in your browser, select **Export**. +1. To export the leaderboard programmatically, select **Python** or **cURL**, then select **Copy** and run the script or command. diff --git a/snippets/_includes/multi-tenant-cloud-only.mdx b/snippets/_includes/multi-tenant-cloud-only.mdx index fff4b14f7b..74baa0aba4 100644 --- a/snippets/_includes/multi-tenant-cloud-only.mdx +++ b/snippets/_includes/multi-tenant-cloud-only.mdx @@ -1 +1 @@ -Currently available only in [W&B Multi-tenant Cloud](/platform/hosting/#wb-multi-tenant-cloud). +This feature is available only in [W&B Multi-tenant Cloud](/platform/hosting/#wb-multi-tenant-cloud). diff --git a/snippets/_includes/object-storage-configuration-intro.mdx b/snippets/_includes/object-storage-configuration-intro.mdx index b8c2e8534e..c5d01eb4e3 100644 --- a/snippets/_includes/object-storage-configuration-intro.mdx +++ b/snippets/_includes/object-storage-configuration-intro.mdx @@ -1,7 +1,7 @@ ### Configure W&B to use your bucket -Once your bucket is provisioned, configure W&B to use it: +After your bucket is provisioned, configure W&B to use it: -- **Self-Managed deployments**: See the [Operator configuration guide](/platform/hosting/self-managed/operator#object-storage-bucket) for Helm values configuration -- **Dedicated Cloud**: Configure via the [System Console](/platform/hosting/iam/sso#system-console) -- **Multi-tenant Cloud**: Configure when creating or editing a team +- **Self-Managed deployments**: Configure Helm values using the [Operator configuration guide](/platform/hosting/self-managed/operator#object-storage-bucket). +- **Dedicated Cloud**: Configure through the [System Console](/platform/hosting/iam/sso#system-console). +- **Multi-tenant Cloud**: Configure when creating or editing a team. diff --git a/snippets/_includes/org-service-account-create.mdx b/snippets/_includes/org-service-account-create.mdx index 55fdfb4e3b..a34eec28a4 100644 --- a/snippets/_includes/org-service-account-create.mdx +++ b/snippets/_includes/org-service-account-create.mdx @@ -1,4 +1,4 @@ -To create a new organization-scoped service account and API key: +To create an organization-scoped service account and API key: 1. Log in to W&B, click your user profile icon, then: - **Dedicated Cloud** or **Self-Managed**: Click **Organization Dashboard**, then click **Service Accounts**. @@ -6,7 +6,7 @@ To create a new organization-scoped service account and API key: 2. Click **Create service account**. 3. Provide a name and select a default team. 4. Click **Create**. -5. Find the service account you just created. +5. Find the service account you created. 6. Click the **action ()** menu, then click **Create API key**. 7. Provide a name for the API key, then click **Create**. 8. Copy the API key and store it securely. diff --git a/snippets/_includes/pinned-and-baseline-runs/pin-runs-condensed.mdx b/snippets/_includes/pinned-and-baseline-runs/pin-runs-condensed.mdx index e82d9f4fb8..1cfe6b40ea 100644 --- a/snippets/_includes/pinned-and-baseline-runs/pin-runs-condensed.mdx +++ b/snippets/_includes/pinned-and-baseline-runs/pin-runs-condensed.mdx @@ -1,7 +1,9 @@ -Pin up to 6 runs to keep them easily accessible at the top of your workspace. If you have a baseline run, you can pin up to 5 runs because the baseline is implicitly pinned. To hide a pinned or baseline run, click the icon. To show a hidden run, click the icon. Pinned runs appear at the top of the run selector with a circular pin icon, separated from other runs by a visual divider. +Pin up to six runs to keep them accessible at the top of your workspace. If you have a baseline run, you can pin up to five runs because the baseline is implicitly pinned. Pinned runs appear at the top of the run selector with a circular pin icon, separated from other runs by a visual divider. + +To hide a pinned or baseline run, click the icon. To show a hidden run, click the icon. To pin a run: -1. Navigate to your workspace. +1. Go to your workspace. 2. In the run selector or runs table, find the run you want to pin. 3. Click the **action ()** menu, then select **Pin run**. diff --git a/snippets/_includes/private-preview-feature.mdx b/snippets/_includes/private-preview-feature.mdx index bc089b0729..ba2cd8b0df 100644 --- a/snippets/_includes/private-preview-feature.mdx +++ b/snippets/_includes/private-preview-feature.mdx @@ -1 +1 @@ -This feature is in **private preview**, available by invitation only. To request enrollment, contact [support](mailto:support@wandb.com) or your AISE. \ No newline at end of file +This feature is in private preview, available by invitation only. To request enrollment, contact [support](mailto:support@wandb.com) or your Account Solutions Engineer (AISE). \ No newline at end of file diff --git a/snippets/_includes/project-visibility-settings.mdx b/snippets/_includes/project-visibility-settings.mdx index 8225262c28..9d2d3a7cd8 100644 --- a/snippets/_includes/project-visibility-settings.mdx +++ b/snippets/_includes/project-visibility-settings.mdx @@ -3,13 +3,15 @@ 1. Choose a new value for **Project visibility**: - **Team** (default): Only your team can view and edit the project. - - **Restricted**: Only invited members can access the project, and public access is turned off. + - **Restricted**: Only invited members can access the project, and public access is deactivated. - **Open**: Anyone can submit runs or create reports, but only your team can edit it. Appropriate only for classroom settings, public benchmark competitions, or other non-durable contexts. - **Public**: Anyone can view the project, but only your team can edit it. - If your W&B admins have turned off **Public** visibility, you cannot choose it. Instead, you can share a view-only [W&B Report](/models/reports/collaborate-on-reports#share-a-report), or contact your W&B organization's admins for assistance. + If your W&B admins have deactivated **Public** visibility, you can't choose it. Instead, you can share a view-only [W&B Report](/models/reports/collaborate-on-reports#share-a-report), or contact your W&B organization's admins for help. 1. Click **Save**. - -If you update a project to a more strict privacy setting, you may need to re-invite individual users to restore their ability to access the project. \ No newline at end of file + + +If you update a project to a stricter privacy setting, you might need to re-invite users to restore their access to the project. + \ No newline at end of file diff --git a/snippets/_includes/public-api-use.mdx b/snippets/_includes/public-api-use.mdx index 7ff388ad84..f95c29f75a 100644 --- a/snippets/_includes/public-api-use.mdx +++ b/snippets/_includes/public-api-use.mdx @@ -1 +1 @@ -> Training and fine-tuning models is done elsewhere in [the W&B Python SDK](/models/ref/python). Use the Public API for querying and managing data *after* it has been logged to W&B. +> Train and fine-tune models with [the W&B Python SDK](/models/ref/python). Use the Public API to query and manage data after you log it to W&B. diff --git a/snippets/_includes/rate-limits-defaults-and-notification.mdx b/snippets/_includes/rate-limits-defaults-and-notification.mdx index d8f5d4585c..d47a4306c4 100644 --- a/snippets/_includes/rate-limits-defaults-and-notification.mdx +++ b/snippets/_includes/rate-limits-defaults-and-notification.mdx @@ -7,12 +7,13 @@ The following default limits help maintain platform stability: | Filestream requests per second | 2 | Run | | Run creation requests per second | 80 | Project | -When a limit is exceeded, the W&B SDK returns HTTP response `429`, and the message `HTTP 429: rate limited exceeded` appears in the SDK logs. +When a limit is exceeded, the W&B SDK returns HTTP response `429`, and the message `HTTP 429: rate limit exceeded` appears in the SDK logs. -- Filesystem rate limits never cause logging to crash or fail. When the SDK receives a `429` response on a filestream request, it will back off and retry the rate-limited request as-is, while subsequent updates accumulate. +Rate limits behave differently depending on the type of request: +- Filestream rate limits never cause logging to crash or fail. When the SDK receives a `429` response on a filestream request, it backs off and retries the rate-limited request as-is, while subsequent updates accumulate. - Run creation rate limits block further training. -``` +```text HTTP 429: rate limit exceeded ``` diff --git a/snippets/_includes/release-notes-support-eol-reminder.mdx b/snippets/_includes/release-notes-support-eol-reminder.mdx index 3846b6bf03..f7e4261091 100644 --- a/snippets/_includes/release-notes-support-eol-reminder.mdx +++ b/snippets/_includes/release-notes-support-eol-reminder.mdx @@ -1 +1 @@ -For details, see [Supported Server releases](https://docs.wandb.ai/release-notes/server-releases) and [Archived Server releases](https://docs.wandb.ai/release-notes/server-releases-archived). A W&B Server release is supported for 12 months from its initial release date. As a reminder, customers using [Self-Managed](/platform/hosting/hosting-options/self-managed/) are responsible to upgrade to a supported release in time to maintain support. +W&B supports each W&B Server release for 12 months from its initial release date. If you use [Self-Managed](/platform/hosting/hosting-options/self-managed/), you're responsible for upgrading to a supported release in time to maintain support. For details, see [Supported Server releases](https://docs.wandb.ai/release-notes/server-releases) and [Archived Server releases](https://docs.wandb.ai/release-notes/server-releases-archived). diff --git a/snippets/_includes/self-managed-hardware-requirements.mdx b/snippets/_includes/self-managed-hardware-requirements.mdx index 740eae3e7e..7dcfc1dc66 100644 --- a/snippets/_includes/self-managed-hardware-requirements.mdx +++ b/snippets/_includes/self-managed-hardware-requirements.mdx @@ -1,3 +1,3 @@ -**CPU Architecture**: W&B runs on Intel (x86) CPU architecture only. ARM is not supported. +**CPU Architecture**: W&B runs on Intel (x86) CPU architecture only. W&B doesn't support ARM. -**Sizing**: For CPU, memory, and disk sizing recommendations for Kubernetes nodes and MySQL, see the [Sizing section](/platform/hosting/self-managed/ref-arch/#sizing) in the reference architecture. Requirements vary based on whether you're running Models, Weave, or both. +**Sizing**: For CPU, memory, and disk sizing recommendations for Kubernetes Nodes and MySQL, see the [Sizing section](/platform/hosting/self-managed/ref-arch/#sizing) in the reference architecture. Requirements vary based on whether you're running Models, Weave, or both. diff --git a/snippets/_includes/self-managed-mysql-requirements.mdx b/snippets/_includes/self-managed-mysql-requirements.mdx index 61649d0fc1..e509adce63 100644 --- a/snippets/_includes/self-managed-mysql-requirements.mdx +++ b/snippets/_includes/self-managed-mysql-requirements.mdx @@ -1,10 +1,10 @@ -W&B requires an external MySQL database. +W&B requires an external MySQL database. -For production, W&B strongly recommends using managed database services: +For production deployments, we recommend a managed database service: - [AWS RDS Aurora MySQL](https://aws.amazon.com/rds/aurora/) - [Google Cloud SQL for MySQL](https://cloud.google.com/sql/mysql) - [Azure Database for MySQL](https://azure.microsoft.com/en-us/products/mysql/) -Managed database services provide automated backups, monitoring, high availability, patching, and reduce operational overhead. +Managed database services reduce operational overhead by providing automated backups, monitoring, high availability, and patching. See the [reference architecture](/platform/hosting/self-managed/ref-arch/#mysql) for complete MySQL requirements, including sizing recommendations and configuration parameters. For database creation SQL, see the [bare-metal guide](/platform/hosting/self-managed/operator/#mysql-database). For questions about your deployment's database configuration, contact [support](mailto:support@wandb.com) or your AISE. diff --git a/snippets/_includes/self-managed-networking-requirements.mdx b/snippets/_includes/self-managed-networking-requirements.mdx index f563e33d14..6c086cddc2 100644 --- a/snippets/_includes/self-managed-networking-requirements.mdx +++ b/snippets/_includes/self-managed-networking-requirements.mdx @@ -1,13 +1,12 @@ -For a networked deployment, egress to these endpoints is required during _both_ installation and runtime: -* https://deploy.wandb.ai -* https://charts.wandb.ai -* https://quay.io (used for Prometheus images) +For a networked deployment, egress to the following endpoints is required during _both_ installation and runtime: +* `https://deploy.wandb.ai` +* `https://charts.wandb.ai` +* `https://quay.io` (used for Prometheus images) -Additional container registries may be required depending on your deployment configuration: -- `https://gcr.io` is needed when deploying Bufstream and etcd for Weave online evaluations. +Additional container registries may be required for some deployment configurations. For example, `https://gcr.io` is required when deploying Bufstream and etcd for Weave online evaluations. -To learn about air-gapped deployments, refer to [Kubernetes operator for air-gapped instances](/platform/hosting/self-managed/on-premises-deployments/kubernetes-airgapped). +For air-gapped deployments, see [Kubernetes operator for air-gapped instances](/platform/hosting/self-managed/on-premises-deployments/kubernetes-airgapped). -Access to W&B and to the object storage is required for the training infrastructure and for each system that tracks the needs of experiments. +Your training infrastructure and any system that tracks the needs of experiments require access to W&B and to the object storage. diff --git a/snippets/_includes/self-managed-object-storage-requirements.mdx b/snippets/_includes/self-managed-object-storage-requirements.mdx index 110f4b1ba0..9e4f6d3483 100644 --- a/snippets/_includes/self-managed-object-storage-requirements.mdx +++ b/snippets/_includes/self-managed-object-storage-requirements.mdx @@ -1,16 +1,17 @@ -W&B requires object storage with pre-signed URL and CORS support. +W&B requires object storage with pre-signed URL and Cross-Origin Resource Sharing (CORS) support. -**Recommended storage providers:** -- [Amazon S3](https://aws.amazon.com/s3/): Object storage service offering industry-leading scalability, data availability, security, and performance. +W&B recommends the following storage providers: + +- [Amazon S3](https://aws.amazon.com/s3/): Object storage service offering scalability, data availability, security, and performance. - [Google Cloud Storage](https://cloud.google.com/storage): Managed service for storing unstructured data at scale. -- [Azure Blob Storage](https://azure.microsoft.com/en-us/products/storage/blobs): Cloud-based object storage solution for storing massive amounts of unstructured data. +- [Azure Blob Storage](https://azure.microsoft.com/en-us/products/storage/blobs): Cloud-based object storage for unstructured data at scale. - [CoreWeave AI Object Storage](https://docs.coreweave.com/products/storage/object-storage): High-performance, S3-compatible object storage service optimized for AI workloads. -- Enterprise S3-compatible storage: [MinIO Enterprise (AIStor)](https://www.min.io/product/aistor), [NetApp StorageGRID](https://www.netapp.com/data-storage/storagegrid/), or other enterprise-grade solutions +- Enterprise S3-compatible storage: [MinIO Enterprise (AIStor)](https://www.min.io/product/aistor), [NetApp StorageGRID](https://www.netapp.com/data-storage/storagegrid/), or other enterprise-grade solutions. -MinIO Open Source is in [maintenance mode](https://github.com/minio/minio) with no active development or pre-compiled binaries. For production deployments, W&B recommends using managed object storage services or enterprise S3-compatible solutions such as MinIO Enterprise (AIStor). +MinIO Open Source is in [maintenance mode](https://github.com/minio/minio) with no active development or precompiled binaries. For production deployments, W&B recommends managed object storage services or enterprise S3-compatible solutions such as MinIO Enterprise (AIStor). For detailed bucket provisioning instructions including IAM policies, CORS configuration, and access setup, see the [Bring Your Own Bucket (BYOB) guide](/platform/hosting/data-security/secure-storage-connector). -See the [reference architecture object storage section](/platform/hosting/self-managed/ref-arch/#object-storage) for complete requirements. +For complete requirements, see the [reference architecture object storage section](/platform/hosting/self-managed/ref-arch/#object-storage). diff --git a/snippets/_includes/self-managed-obtain-license.mdx b/snippets/_includes/self-managed-obtain-license.mdx index ce83c32840..f6785cb6bf 100644 --- a/snippets/_includes/self-managed-obtain-license.mdx +++ b/snippets/_includes/self-managed-obtain-license.mdx @@ -1,11 +1,12 @@ -You need a W&B license to deploy W&B Self-Managed. +To obtain a license, follow these steps: -1. If you do not already have a W&B account, create one. -2. If you need an enterprise trial license with support for important security and other enterprise-friendly capabilities, [submit a request](https://wandb.ai/site/for-enterprise/self-hosted-trial) or reach out to your W&B team. +1. If you don't already have a W&B account, create one. +2. If you need an enterprise trial license with support for security and other enterprise capabilities, [submit a request](https://wandb.ai/site/for-enterprise/self-hosted-trial) or contact your W&B team. 3. Otherwise, open the [Deploy Manager](https://deploy.wandb.ai/deploy) to generate a free trial license. The URL redirects you to a **Get a License for W&B Local** form. Provide the following information: - The owner of the license - The deployment type - - A name and optional description for the instance + - Name + - Description (Optional) 4. Click **Generate License Key**. -A page displays with an overview of your deployment along with the license associated with the instance. +A page appears with an overview of your deployment and the license associated with the instance. diff --git a/snippets/_includes/self-managed-redis-requirements.mdx b/snippets/_includes/self-managed-redis-requirements.mdx index c2ced8a9fb..3f5bbb6bf5 100644 --- a/snippets/_includes/self-managed-redis-requirements.mdx +++ b/snippets/_includes/self-managed-redis-requirements.mdx @@ -1,8 +1,8 @@ -W&B depends on a single-node Redis 7.x deployment used by W&B's components for job queuing and data caching. For convenience during testing and development of proofs of concept, W&B Self-Managed includes a local Redis deployment that is not appropriate for production deployments. +W&B depends on a single-node Redis 7.x deployment that W&B's components use for job queuing and data caching. For convenience during testing and development of proofs of concept, W&B Self-Managed includes a local Redis deployment that isn't appropriate for production deployments. For production deployments, W&B can connect to a Redis instance in the following environments: - [AWS Elasticache](https://aws.amazon.com/elasticache/) - [Google Cloud Memory Store](https://cloud.google.com/memorystore?hl=en) - [Azure Cache for Redis](https://azure.microsoft.com/en-us/products/cache) -- Redis deployment hosted in your cloud or on-premise infrastructure +- Redis deployment hosted in your cloud or on-premises infrastructure diff --git a/snippets/_includes/self-managed-ssl-tls-requirements.mdx b/snippets/_includes/self-managed-ssl-tls-requirements.mdx index 154c221f1c..1d671bc51c 100644 --- a/snippets/_includes/self-managed-ssl-tls-requirements.mdx +++ b/snippets/_includes/self-managed-ssl-tls-requirements.mdx @@ -1,8 +1,8 @@ -W&B requires a valid signed SSL/TLS certificate for secure communication between clients and the server. SSL/TLS termination must occur on the ingress/load balancer. The W&B Server application does not terminate SSL or TLS connections. +W&B requires a valid signed SSL/TLS certificate for secure communication between clients and the server. The ingress or load balancer must terminate SSL/TLS. The W&B Server application doesn't terminate SSL or TLS connections. -**Important**: W&B does not support self-signed certificates and custom CAs. Using self-signed certificates will cause challenges for users and is not supported. +**Important**: W&B doesn't support self-signed certificates or custom certificate authorities (CAs). Self-signed certificates cause problems for users. -If possible, using a service like [Let's Encrypt](https://letsencrypt.org) is a great way to provide trusted certificates to your load balancer. Services like Caddy and Cloudflare manage SSL for you. +If possible, use a service like [Let's Encrypt](https://letsencrypt.org) to provide trusted certificates to your load balancer. Services like Caddy and Cloudflare manage SSL for you. -If your security policies require SSL communication within your trusted networks, consider using a tool like Istio and [side car containers](https://istio.io/latest/docs/reference/config/networking/sidecar/). +If your security policies require SSL communication within your trusted networks, consider a tool like Istio with [sidecar containers](https://istio.io/latest/docs/reference/config/networking/sidecar/). diff --git a/snippets/_includes/self-managed-verify-installation.mdx b/snippets/_includes/self-managed-verify-installation.mdx index b8bd919a34..3b7e01dee8 100644 --- a/snippets/_includes/self-managed-verify-installation.mdx +++ b/snippets/_includes/self-managed-verify-installation.mdx @@ -1,34 +1,38 @@ -To verify the installation, W&B recommends using the [W&B CLI](/models/ref/cli/). The verify command executes several tests that verify all components and configurations. +To verify the installation, use the [W&B CLI](/models/ref/cli/). The `wandb verify` command runs several tests that confirm all components and configurations work as expected. -This step assumes that the first admin user account is created with the browser. +This procedure assumes that you have already created the first admin user account in the browser. -Follow these steps to verify the installation: +Follow these steps: 1. Install the W&B CLI: -```bash -pip install wandb -``` -2. Log in to W&B: -```bash -wandb login --host=https://YOUR_DNS_DOMAIN -``` + ```bash + pip install wandb + ``` -For example: -```bash -wandb login --host=https://wandb.company-name.com -``` +2. Log in to W&B, replacing `[YOUR-DNS-DOMAIN]` with the DNS domain of your deployment: + + ```bash + wandb login --host=https://[YOUR-DNS-DOMAIN] + ``` + + For example: + + ```bash + wandb login --host=https://wandb.company-name.com + ``` 3. Verify the installation: -```bash -wandb verify -``` -A successful installation and fully working W&B deployment shows the following output: + ```bash + wandb verify + ``` + +A successful installation and working W&B deployment shows the following output: -```console +```text Default host selected: https://wandb.company-name.com Find detailed logs for this test at: /var/folders/pn/b3g3gnc11_sbsykqkm3tx5rh0000gp/T/tmpdtdjbxua/wandb Checking if logged in...................................................✅ @@ -42,4 +46,4 @@ Checking logged metrics, saving and downloading a file..................✅ Checking artifact save and download workflows...........................✅ ``` -Contact W&B Support if you encounter errors. +If you encounter errors, contact W&B Support. diff --git a/snippets/_includes/service-account-api-key-create-additional-single-tenant.mdx b/snippets/_includes/service-account-api-key-create-additional-single-tenant.mdx index 54dbcf3465..83556d765d 100644 --- a/snippets/_includes/service-account-api-key-create-additional-single-tenant.mdx +++ b/snippets/_includes/service-account-api-key-create-additional-single-tenant.mdx @@ -1,10 +1,10 @@ To create an API key owned by a service account: -1. Navigate to the **Service Accounts** tab in your team or organization settings. +1. In your team or organization settings, go to the **Service Accounts** tab. 2. Find the service account in the list. 3. Click the **action ()** menu, then click **Create API key**. 4. Provide a name for the API key, then click **Create**. -5. Copy the displayed API key immediately and store it securely. +5. Copy the displayed API key immediately and store it securely. W&B shows the key only once and can't retrieve it later. 6. Click **Done**. You can create multiple API keys for a single service account to support different environments or workflows. diff --git a/snippets/_includes/service-account-api-key-delete.mdx b/snippets/_includes/service-account-api-key-delete.mdx index d82ff33cc1..51ee9683c4 100644 --- a/snippets/_includes/service-account-api-key-delete.mdx +++ b/snippets/_includes/service-account-api-key-delete.mdx @@ -1,5 +1,5 @@ To delete an API key owned by an organization or team service account: -1. Go to [Organization settings](https://wandb.ai/account-settings/), then click **API Keys**. -2. Find the API key. The list includes all API keys owned by organization and team service accounts. You can search or filter by key name or ID, and you can sort by any column. -3. Click the delete button. +1. Go to [Organization settings](https://wandb.ai/account-settings/) > **API Keys**. +2. Find the API key. The list includes all API keys owned by organization and team service accounts. You can search or filter by key name or ID, and sort by any column. +3. Click the **Delete** button. diff --git a/snippets/_includes/service-account-benefits.mdx b/snippets/_includes/service-account-benefits.mdx index a311d4b8df..6b2806799d 100644 --- a/snippets/_includes/service-account-benefits.mdx +++ b/snippets/_includes/service-account-benefits.mdx @@ -1,6 +1,6 @@ -Key benefits of service accounts: -- **No license consumption**: Service accounts do not consume user seats or licenses -- **Dedicated API keys**: Secure credentials for automated workflows -- **User attribution**: Optionally attribute automated runs to human users -- **Enterprise-ready**: Built for production automation at scale -- **Delegated operations**: Service accounts operate on behalf of the user or organization that creates them \ No newline at end of file +Service accounts provide the following key benefits: +- **No license consumption**: Service accounts don't consume user seats or licenses. +- **Dedicated API keys**: Secure credentials for automated workflows. +- **User attribution**: Optionally attribute automated runs to human users. +- **Enterprise-ready**: Built for production automation at scale. +- **Delegated operations**: Service accounts operate on behalf of the user or organization that creates them. \ No newline at end of file diff --git a/snippets/_includes/team-service-account-create.mdx b/snippets/_includes/team-service-account-create.mdx index 9f8ef5937a..98bcb8d08f 100644 --- a/snippets/_includes/team-service-account-create.mdx +++ b/snippets/_includes/team-service-account-create.mdx @@ -3,10 +3,10 @@ To create a new team-scoped service account and API key: 1. In your team's settings, click **Service Accounts**. 2. Click **New Team Service Account**. 3. Provide a name for the service account. -4. Set Authentication Method to **Generate API key** (default). If you select **Federated Identity**, the service account cannot own API keys. +4. Set **Authentication Method** to **Generate API key** (default). If you select **Federated Identity**, the service account can't own API keys. 5. Click **Create**. 6. Find the service account you just created. -7. Click the **action ()** menu, then click **Create API key**. +7. Click the **action** () menu, then click **Create API key**. 8. Provide a name for the API key, then click **Create**. 9. Copy the API key and store it securely. 10. Click **Done**. diff --git a/snippets/_includes/weave-quickstart-prereqs.mdx b/snippets/_includes/weave-quickstart-prereqs.mdx index 6a95e7cd90..d69373d2a9 100644 --- a/snippets/_includes/weave-quickstart-prereqs.mdx +++ b/snippets/_includes/weave-quickstart-prereqs.mdx @@ -1,8 +1,10 @@ ## Prerequisites -- A [W&B account](https://wandb.ai/signup) -- Python 3.10+ or Node.js 18+ +Before you begin, you must have the following: + +- A [W&B account](https://wandb.ai/signup). +- Python 3.10+ or Node.js 18+. - Required packages installed: - **Python**: `pip install weave openai` - **TypeScript**: `npm install weave openai` -- An [OpenAI API key](https://platform.openai.com/api-keys) set as an environment variable \ No newline at end of file +- An [OpenAI API key](https://platform.openai.com/api-keys) set as an environment variable. \ No newline at end of file