New Product Feature Overview
Q3 2024
Created on August 6|Last edited on August 8
Comment
1. SDK Performance Boost ⚡
What's new in SDK? wandb-core is a new and improved backend for the W&B SDK that is more performant, versatile, and robust. Explore the repo to learn more about the new improvements to our SDK and take full advantage of these features and improvements today.
Highlights:
- Logging Performance: Up to 88% performance improvements with multiple processes.
- Improved Table Logging: Up to 40% faster tables logging.
- Faster Startups & Shutdowns: Up to 36% faster times.
- Enhanced Artifact Handling: Up to 33% faster uploads and 27% faster retrievals.
- Faster Offline Sync.
How to enable? Upgrade to >=wandb v0.17.5+ and add wandb.require("core") to your scripts for improved logging performance.
2. W&B Run Rewind ⏪
What is Rewind? Rewind allows you to "rewind" a run’s history to a specific point, enabling truncation or modification without losing original data.
This compliments the run forking capability below, allowing you to correct or modify the run history itself!
Highlights:
- History Truncation: Modify run history without losing data. Truncate run history to the rewind point, allowing new data logging and recomputed summary metrics.
- Config Preservation: Preserve and merge original configurations with new ones.
- Run & Artifact Management: Artifacts are associated with the source run that produced them.
- Consistent Management: Maintain consistent run IDs for traceable experiments.
How to enable? Upgrade to >=wandb v0.17.5+
Note: This is not yet supported in the new sdk backend wandb-core
3. W&B Run Forking 🌿
What is Run Forking? Run forking allows you to create a new run, or fork from an existing W&B run, enabling users to explore different parameters or models from a specific step without impacting the original run. This is particularly useful for continuous experimentation and iterative improvements with longer training runs.
This compliments the run rewind capability above, providing more flexibility in managing and experimenting with your runs as opposed to modifying their history!
Highlights:
- History Preservation: Forking preserves the integrity of an experiment appending its history and configuration to the new run, enabling easy comparisons and adjustments.
- Simplified Experience: Forking simplifies testing alternate hypotheses or parameters from any _step, preserving the original source run's integrity.
- Incremental Analysis: Forking offers an incremental view of your training experiment, capturing the essence of each variation and its impact to the full training run.
How to enable? Upgrade to >=wandb v0.17.5+
Note: This is not yet supported in the new sdk backend wandb-core
In the example below, multiple runs are forked from the run_fork_parent run at step 60, preserving the original run's state. This approach allows for iterative experimentation while maintaining a consistent baseline. By forking at a particular step, you ensure that the new runs inherit all the parameters, and data up to that point, enabling focused modifications and optimizations from a known state.
Run set
10
4. W&B Workspace API 🚀
What is W&B Workspace API? This update allows for the programmatic creation and manipulation of W&B Workspaces, enhancing users' workflow and productivity.
Highlights:
- Customizable Workspaces: Define, create, and customize workspaces with specific layouts, colors, and sections.
- Editable Views: Load, modify, and save changes to existing workspaces or create new views.
- Run Management: Programmatically filter, group, and sort runs, and customize their appearance.
For more details and code examples, check out the documentation and this tutorial. This is an optional Python package and can be installed with: pip install wandb[workspaces] using >=wandb v0.17.5+.
5. W&B Charts Full Fidelity plots 📈
What are Full Fidelity Plots? A plot point aggregation method, which plots average values while preserving minimum and maximum values across x-axis point buckets. Providing a complete and precise representation of training runs, allowing users to see detailed outliers and zoom in without losing data integrity.
Highlights:
- Show all outliers/spikes: Significantly enhance the visibility of training metrics and results, especially for long training runs or large training jobs. Users can now visualize outliers and spikes alongside all other metrics in a single, comprehensive line plot without random sampling.
- Improved line detail: Aggregates data along the x-axis, showing minimum, maximum, and average values for a complete representation.
- Full fidelity: Spot outliers and zoom into high-fidelity details without data loss.
Run set
21
Add a comment