You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: content/guides/models/track/log/distributed-training.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -105,7 +105,7 @@ Parameters prefixed by `x_` (such as `x_label`) are in public preview. Create a
105
105
{{% /alert %}}
106
106
107
107
{{% alert %}}
108
-
To track multiple processes to a single run, you must have W&B Python SDK version `v0.19.5` or newer.
108
+
To track multiple processes to a single run, you must have W&B Python SDK version `v0.19.9` or newer.
109
109
{{% /alert %}}
110
110
111
111
In this approach you use a primary node and one or more worker nodes. Within the primary node you initialize a W&B run. For each worker node, initialize a run using the run ID used by the primary node. During training each worker node logs to the same run ID as the primary node. W&B aggregates metrics from all nodes and displays them in the W&B App UI.
@@ -268,4 +268,4 @@ For optimal experience we do recommend you upgrade to the latest version.
268
268
269
269
**W&B SDK 0.12.4 and below**
270
270
271
-
Set the `WANDB_START_METHOD` environment variable to `"thread"` to use multithreading instead if you use a W&B SDK version 0.12.4 and below.
271
+
Set the `WANDB_START_METHOD` environment variable to `"thread"` to use multithreading instead if you use a W&B SDK version 0.12.4 and below.
Copy file name to clipboardExpand all lines: content/guides/quickstart.md
+6Lines changed: 6 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -10,6 +10,10 @@ weight: 2
10
10
---
11
11
Install W&B to track, visualize, and manage machine learning experiments of any size.
12
12
13
+
{{% alert %}}
14
+
Are you looking for information on W&B Weave? See the [Weave Python SDK quickstart](https://weave-docs.wandb.ai/quickstart) or [Weave TypeScript SDK quickstart](https://weave-docs.wandb.ai/reference/generated_typescript_docs/intro-notebook).
15
+
{{% /alert %}}
16
+
13
17
## Sign up and create an API key
14
18
15
19
To authenticate your machine with W&B, generate an API key from your user profile or at [wandb.ai/authorize](https://wandb.ai/authorize). Copy the API key and store it securely.
@@ -41,6 +45,7 @@ pip install wandb
41
45
```
42
46
```python
43
47
import wandb
48
+
44
49
wandb.login()
45
50
```
46
51
@@ -122,3 +127,4 @@ Explore more features of the W&B ecosystem:
122
127
4. Automate hyperparameter searches and optimize models with [W&B Sweeps]({{< relref "/guides/models/sweeps/">}}).
123
128
5. Analyze runs, visualize model predictions, and share insights on a [central dashboard]({{< relref "/guides/models/tables/">}}).
124
129
6. Visit [W&B AI Academy](https://wandb.ai/site/courses/) to learn about LLMs, MLOps, and W&B Models through hands-on courses.
130
+
7. Visit the [official W&B Weave documentation](https://weave-docs.wandb.ai/) to learn how to track track, experiment with, evaluate, deploy, and improve your LLM-based applications using Weave.
Are you looking for the official Weave documentation? Head over to [https://weave-docs.wandb.ai/](https://weave-docs.wandb.ai/).
11
+
{{% /alert %}}
10
12
11
-
Weave is a lightweight toolkit for tracking and evaluatingLLM applications. Use W&B Weave to visualize and inspect the execution flow of your LLMs, analyze the inputs and outputs of your LLMs, view the intermediate results and securely store and manage your prompts and LLM chain configurations.
13
+
W&B Weave is a framework for tracking, experimenting with, evaluating, deploying, and improving LLM-based applications. Designed for flexibility and scalability, Weave supports every stage of your LLM application development workflow:
-**Tracing & Monitoring**: Track LLM calls and application logic to debug and analyze production systems.
16
+
-**Systematic Iteration**: Refine and iterate on prompts, datasets and models.
17
+
-**Experimentation**: Experiment with different models and prompts in the LLM Playground.
18
+
-**Evaluation**: Use custom or pre-built scorers alongside our comparison tools to systematically assess and enhance application performance.
19
+
-**Guardrails**: Protect your application with safeguards for content moderation, prompt safety, and more.
14
20
15
-
With W&B Weave, you can:
16
-
* Log and debug language model inputs, outputs, and traces
17
-
* Build rigorous, apples-to-apples evaluations for language model use cases
18
-
* Organize all the information generated across the LLM workflow, from experimentation to evaluations to production
21
+
## Get started with Weave
19
22
20
-
{{% alert %}}
21
-
Looking for Weave docs? See the [W&B Weave Docs](https://weave-docs.wandb.ai/).
22
-
{{% /alert %}}
23
+
Are you new to Weave? Set up and start using Weave with the [Python quickstart](https://weave-docs.wandb.ai/quickstart) or [TypeScript quickstart](https://weave-docs.wandb.ai/reference/generated_typescript_docs/intro-notebook).
24
+
25
+
## Advanced guides
23
26
24
-
## How to get started
25
-
Depending on your use case, explore the following resources to get started with W&B Weave:
27
+
Learn more about advanced topics:
26
28
27
-
*[Quickstart: Track inputs and outputs of LLM calls](https://wandb.github.io/weave/quickstart)
28
-
*[Build an Evaluation pipeline tutorial](https://wandb.github.io/weave/tutorial-eval)
29
-
*[Model-Based Evaluation of RAG applications tutorial](https://wandb.github.io/weave/tutorial-rag)
29
+
-[Integrations](https://weave-docs.wandb.ai/guides/integrations/): Use Weave with popular LLM providers, local models, frameworks, and third-party services.
30
+
-[Cookbooks](https://weave-docs.wandb.ai/reference/gen_notebooks/intro_notebook): Build with Weave using Python and TypeScript. Tutorials are available as interactive notebooks.
31
+
-[W&B AI Academy](https://www.wandb.courses/pages/w-b-courses): Build advanced RAG systems, improve LLM prompting, fine-tune LLMs, and more.
Flexible and lightweight building block for dataset and model versioning.
8
8
@@ -44,6 +44,7 @@ begin with `add`. Once the artifact has all the desired files, you can call
44
44
|`digest`| The logical digest of the artifact. The digest is the checksum of the artifact's contents. If an artifact has the same digest as the current `latest` version, then `log_artifact` is a no-op. |
45
45
|`entity`| The name of the entity of the secondary (portfolio) artifact collection. |
46
46
|`file_count`| The number of files (including references). |
47
+
|`history_step`| The nearest step at which history metrics were logged for the source run of the artifact. |
47
48
|`id`| The artifact's ID. |
48
49
|`manifest`| The artifact's manifest. The manifest lists all of its contents, and can't be changed once the artifact has been logged. |
49
50
|`metadata`| User-defined artifact metadata. Structured data associated with the artifact. |
@@ -69,7 +70,7 @@ begin with `add`. Once the artifact has all the desired files, you can call
0 commit comments