Week of October 28th
Image by author.

Week of October 28th

TL;DR:


Hamilton Release Highlights:

Hamilton Framework == 1.82.0

In-memory cache

If you're iterating in a notebook, you might not want to have everything write to disk on every execution. Instead, you could just hold everything in memory. This is what the new InMemoryResultStore and InMemoryMetadataStore implementations do. They're easy to switch to:

from hamilton.caching.stores.memory import InMemoryMetadataStore, InMemoryResultStore

dr = (
    driver.Builder()
    .with_modules(dataflow_module)
    .with_cache(
        result_store=InMemoryResultStore(),
        metadata_store=InMemoryMetadataStore(),
    )
    .build()
)        

Then any subsequent runs will have everything cached in memory. But what happens when you want to exit your notebook? Well you can persist what was in memory to disk:

from hamilton.caching.stores.sqlite import SQLiteMetadataStore
from hamilton.caching.stores.file import FileResultStore

path = "./.persisted_cache"
on_disk_results = FileResultStore(path=path)
on_disk_metadata = SQLiteMetadataStore(path=path)

# use driver you were using above
dr.cache.result_store.persist_to(on_disk_results)
dr.cache.metadata_store.persist_to(on_disk_metadata)        

Then we can also rehydrate from persisted results, with the load_from() functions:

dr = (
    driver.Builder()
    .with_modules(dataflow_module)
    .with_cache(
        result_store=InMemoryResultStore.load_from(
            on_disk_results,
            metadata_store=on_disk_metadata,
        ),
        metadata_store=InMemoryMetadataStore.load_from(on_disk_metadata),
    )
    .build()
)        

Thanks to Thierry Jean for driving this feature. For more details see this notebook tutorial.


Burr Release Highlights

Burr == 0.33.0rc0 (release candidate)

Framework level Parallelism (preview)

With this release candidate, we're enabling Burr to run multiple actions in parallel. Each parallel "edge" can contain one or more actions, and different "edges" can have different actions. This is useful when:

  • Trying different prompts with an LLM
  • Trying a prompt with different LLMs
  • Trying multiple prompts with multiple LLMs
  • Running multiple tools in parallel then combining/selecting from the result
  • Do semantic search and web search simultaneously for information retrieval

And more! Just like Burr in general, these concepts are generic and can be applied to non-LLM applications. So we're excited to push a release candidate for early feedback on the API and UX. So if this interests you, you can read more here, and otherwise we'd love your reactions and feedback!


Office Hours & Meetup

Hamilton Meet up: Our next meet-up will be December. Want to present? Reach out. Otherwise join/sign-up here.

Hamilton Office Hours: They happen most Tuesday 9:30am PT - 10:30am PT.

Join our slack for the link.

Burr Office Hours: They happen most Wednesdays 9:30am PT - 10:3am PT.

Join our discord for the weekly link.


MLOps World & Generative AI World Summit 2024

This November is the annual MLOps & Generative AI World summit. It's in Austin, Texas. I went last year and had a great series of conversations with practitioners. If you can make it, I'd recommend attending.

For those that don't know, the goal of the summit/conference, organized by the Toronto Machine Learning Society (TMLS) , is to help companies put more machine learning and AI into production environments, effectively, responsibly, and efficiently.

Whether you're working towards a live production deployment, or currently working in production, this is a conference geared towards the gathering of like minded individuals to help share practical knowledge to help you on your journey.

Some of the talk tracks this year:

  • Real World Case Studies
  • Business & Strategy
  • Technical & Research (levels 1-7)
  • Workshops (levels 1-7) <-- I'll be doing one!
  • In-person coding sessions

GenAI for SWEs Workshop

Together with Hugo Bowne-Anderson I will be hosting a workshop for software engineers on some first principles for delivering GenAI applications. More details to follow.

I'll also be running a community table on Hamilton & Burr, and "reliable AI" best practices.

Discount for Passes

If you'd like to attend, you can use the code DAGWORKS150 to get $150 off all passes. If you're going, send me a note, I'd love to meet-up.

Conference Details

When: 9AM ET on Thursday, November 7th to 5PM ET on Friday, November 8th 2024 Where: Renaissance Austin Hotel, 9721 Arboretum Boulevard, Austin, TX. MAP.

Need more convincing? Watch this video.

Want a free pass to our Workshop?

If you're a SWE and want to know more about GenAI? Comment on David Scharbach 's post and you could win a free pass to our workshop -> comment here.



Running a Maven course on Building GenAI Applications

I'm excited to partner with Hugo Bowne-Anderson and Maven to build out a course to help ground the principles required to ship GenAI applications.

But, we want to make sure it is relevant! So if you can spare 5-10 minutes, we'd love your feedback to help shape the course: https://maven.com/forms/b83ce4


Blog Post: Build LLM agents faster with Haystack + Burr!

We're excited to showcase how versatile Burr is and how you can utilize an existing framework's "off-the-shelf" components in conjunction with the observability and state management that Burr provides.


In the Wild:

Burr at DataForAI meetup in SF this month

Burr was at this month's DataForAI meetup. Recording to come, while here's the notebook that I presented.

Hamilton at DataForAI meetup

While I was at the meet-up, Hamilton showed up in the OpenLineage talk. It showed up because Hamilton is an open lineage producer!


The star logo in the wild!

From our friends: DLTHub event

Our friends at dltHub are hostingl a meet-up next week in San Francisco. It's going to be a great event, and we'd love for you join if you're around. Sign-up here:


要查看或添加评论,请登录

社区洞察

其他会员也浏览了