100 indexes is still fairly useless, or I'm not understanding how they should be used. Any normal ap

100 indexes is still fairly useless, or I'm not understanding how they should be used. Any normal app will have easily thousands+ users and thousands+ files. What am I supposed to use as index? maybe file/type but doesn't really help much. Metadata search is a must.

cosbgnOP•10/4/23, 12:21 PM

Just in case my message came out to strong, I love cloudflare

and the amazign work you are doing!! I actually love it that much that I'm trying to fully migrate everything to it! Being able to store vectors here it would be amazing

Ccosbgn 100 indexes is still fairly useless, or I'm not understanding how they should be...

Matt Silverlock•10/4/23, 12:44 PM

It’s not meant as a replacement. Some folks want more indexes for isolating different use-cases, and 10 was just where we started.

Ccosbgn 100 indexes is still fairly useless, or I'm not understanding how they should be...

Matt Silverlock•10/4/23, 12:45 PM

Hold tight on the metadata filtering. We’re working on it

nclevenger•10/4/23, 3:23 PM

Does anyone know what this error means: VECTOR_QUERY_ERROR: Status + 500VECTOR_QUERY_ERROR: Status + 500? I was running the exact query successfully earlier this morning, but now it is erroring.

Nnclevenger Does anyone know what this error means: `VECTOR_QUERY_ERROR: Status + 500`? I w...

Jerome•10/4/23, 3:26 PM

This means internal error, I can look in the internal log

JJerome This means internal error, I can look in the internal log

Isaac McFadyen•10/4/23, 3:27 PM

Just a heads-up that you can type /cfrole/cfrole to get an orange Cloudflare role in the server

IIsaac McFadyen Just a heads-up that you can type `/cfrole` to get an orange Cloudflare role in ...

Jerome•10/4/23, 3:27 PM

yeah

JJerome This means internal error, I can look in the internal log

nclevenger•10/4/23, 3:28 PM

account id a826340b3b93189c9ebb7c0eaeba3c46 index blogs

IIsaac McFadyen Just a heads-up that you can type `/cfrole` to get an orange Cloudflare role in ...

Jerome•10/4/23, 3:28 PM

thx

JJerome 💥 thx

Isaac McFadyen•10/4/23, 3:28 PM

No problem!

IIsaac McFadyen Just a heads-up that you can type `/cfrole` to get an orange Cloudflare role in ...

Old Man Maker•10/4/23, 3:52 PM

What does that mean or refer to?

OOld Man Maker What does that mean or refer to?

Jerome•10/4/23, 3:53 PM

it's a command to mark one's discord account as belonging to an employee of cloudflare

Old Man Maker•10/4/23, 3:54 PM

oh I see

Old Man Maker•10/4/23, 3:54 PM

well now I'm just jealous

Jerome•10/4/23, 4:07 PM

@oldmanmeta have you seen my DM ?

Jerome•10/4/23, 4:11 PM

I mean: @nclevenger, have you seen my DM>

JJerome I mean: @nclevenger, have you seen my DM>

nclevenger•10/4/23, 4:23 PM

no ... I don't think i see it?

JJerome I mean: @nclevenger, have you seen my DM>

nclevenger•10/4/23, 4:25 PM

i found it - thanks!

Old Man Maker•10/4/23, 4:47 PM

I'm getting the same error as @nclevenger

VECTOR_QUERY_ERROR: Status + 404

VECTOR_QUERY_ERROR: Status + 404

Old Man Maker•10/4/23, 4:47 PM

just started happening

Old Man Maker•10/4/23, 4:54 PM

Account ID: 43310a46ee87ca6683a51510f7a57241

Old Man Maker•10/4/23, 4:55 PM

Can some please take a look for me.
The vectors are being created in the query correctly, calling <vectorize>.query() is failing

Old Man Maker•10/4/23, 4:56 PM

the index is ais-vec-bakedcontent-index

OOld Man Maker I'm getting the same error as @nclevenger ``` VECTOR_QUERY_ERROR: Status + 404...

Jerome•10/4/23, 5:08 PM

looking into it

JJerome looking into it

Old Man Maker•10/4/23, 5:09 PM

Thanks Sir

KKarl Hey, i'm playing around with vectorize and workers AI. Really impressed so far....

Matt Silverlock•10/5/23, 10:15 AM

You’d want some form of metadata filtering (we are working on it) for that.

MMatt Silverlock It’s not meant as a replacement. Some folks want more indexes for isolating diff...

Zak @ insertchat.com•10/5/23, 11:24 AM

Yup, that's my case. Do you see any advantages/disadvantages to using one big index with all user vectors + metadata filtering by a user id vs. one index per user?

ZZak @ insertchat.com Yup, that's my case. Do you see any advantages/disadvantages to using one big in...

Matt Silverlock•10/5/23, 11:30 AM

It will depend on your index size. If the count of vectors per user is very small, and will remain so, a larger single index will remain “hot”

Zak @ insertchat.com•10/5/23, 11:31 AM

Oh I see, there is a notion of hot/cold storage involved. What's your definition of small? Millions? Tens of millions? 1 billion?

ZZak @ insertchat.com Oh I see, there is a notion of hot/cold storage involved. What's your definition...

Matt Silverlock•10/5/23, 11:33 AM

per user or per index? Small would be thousands per user - over hundreds of thousands of users -> an index in the millions.

Zak @ insertchat.com•10/5/23, 11:34 AM

I don't have more then a million vector per user

Matt Silverlock•10/5/23, 11:34 AM

To step back a second: it really doesn’t matter that much, and it probably doesn’t make a lot of sense to over engineer this right now

Zak @ insertchat.com•10/5/23, 11:34 AM

The total in my db right now is between 5 and 10 millions vectors

Matt Silverlock•10/5/23, 11:35 AM

That’s fine.

Matt Silverlock•10/5/23, 11:36 AM

Write your raw embeddings to R2. If you want to change your approach, it’s much easier to repopulate an index vs a traditional RDBMS

Zak @ insertchat.com•10/5/23, 11:39 AM

Will dump a few million vectors to Vectorize once we have filtering to see how it performs with a user id filter, if it's good then it's a go for production, and will see if we change it later to multiple indexes

Zak @ insertchat.com•10/5/23, 12:35 PM

Can we also expect a way to CRUD a vector (row) directly by its id/uid?

ZZak @ insertchat.com Can we also expect a way to CRUD a vector (row) directly by its id/uid?

Jerome•10/5/23, 12:45 PM

there is getByIdsgetByIds and deleteByIdsdeleteByIds operations; insertinsert and upsertupsert take the id you provide as well

Zak @ insertchat.com•10/5/23, 12:46 PM

I'm an API user, don't have those (I think)

Zak @ insertchat.com•10/5/23, 1:31 PM

Hopefully, API users will get some love from Cloudflare x)

prmptCoder•10/5/23, 3:14 PM

Is it possible to train or fine-tune the text-generation-model distilbert-sst-2-int8 and deploy@everyone

PprmptCoder Is it possible to train or fine-tune the text-generation-model distilbert-sst-2-...

Old Man Maker•10/5/23, 3:23 PM

AFAIK - model hosting with Hugging Face is on the road map to be exposed by cloudflare, so you should be able to at some point

Andrew Mayne•10/6/23, 12:07 AM

Hey folks! Two questions:

1. Is there some knack to saving meta data? Everything else saves in this example except the metadata:

const embedding = await getOpenAIEmbedding(description, env.OPENAI_KEY);
await env.VECTORIZE.upsert([{ id: promptId, values: embedding, metadata:{ title:"Updated title", description: "Description"} }]);

console.log("Item vector added:", { title: title, description: description })

2. Everytime I update from the CLI my vector binding are erased. Adding this to the TOML doesn't seem to help:

vectorize = [
{ binding = "VECTORIZE", index_name = "my-index" }
]

Or:

[[vectorize]]
binding = "VECTORIZE"
index_name = "my-index"

AAndrew Mayne Hey folks! Two questions: 1. Is there some knack to saving meta data? Everythin...

Old Man Maker•10/6/23, 2:48 AM

Keep in mind that metadata as a type may change in future if filtering by metadata is added.

At the moment, this is how I create it:

//I defined a new object type for the metadata - MetadataSchema 
let newMetaData:MetadataSchema ={
    d1Key: keyValue,
    prjRef: projectRef,
    accRef: accountRef,
    parentRef:parentRef
}

//Stuff the object in there
const metaRecord:Record<string,VectorizeVectorMetadata>= { 
    "data": JSON.stringify(newMetaData)
};

//Add to the vector
let tempVec:VectorizeVector ={
    id:id,
    values: data[0], //the embeddings
    namespace:<The Namespace>,
    metadata:metaRecord
}

//I defined a new object type for the metadata - MetadataSchema 
let newMetaData:MetadataSchema ={
    d1Key: keyValue,
    prjRef: projectRef,
    accRef: accountRef,
    parentRef:parentRef
}

//Stuff the object in there
const metaRecord:Record<string,VectorizeVectorMetadata>= { 
    "data": JSON.stringify(newMetaData)
};

//Add to the vector
let tempVec:VectorizeVector ={
    id:id,
    values: data[0], //the embeddings
    namespace:<The Namespace>,
    metadata:metaRecord
}

100 indexes is still fairly useless, or I'm not understanding how they should be used. Any normal ap

Similar Threads

Similar Threads

Similar Threads