trulens VS LANDrop

Retrieval using a single vector is called dense passage retrieval (DPR), because an entire passage (dozens to hundreds of tokens) is encoded as a single vector. ColBERT instead encodes a vector-per-token, where each vector is influenced by surrounding context. This leads to meaningfully better results; for example, here’s ColBERT running on Astra DB compared to DPR using openai-v3-small vectors, compared with TruLens for the Braintrust Coda Help Desk data set. ColBERT easily beats DPR at correctness, context relevance, and groundedness.

12. TruLens by TruEra | Github | tutorial

3) Hallucination is probably the biggest problem we solve for. To do evals for hallucination, we typically see our users use a combination of groundedness (does the context support the LLM response) and context relevance (is the retrieved context relevant to the query). There's also a bunch more for the evaluations you mentioned (moderation models, sentiment, usefulness, etc.) and it's pretty easy to add custom evals.

Also - my hot take is that gpt-3.5 is good enough for evals (sometimes better) than gpt-4 if you give the LLM enough instructions on how to do the eval.

website: https://www.trulens.org/

The team at TruEra recently released an open source project for evaluation & tracking of LLM applications called TruLens-Eval. We’ve specifically targeted retrieval-augmented QA as a core use case and so far we’ve seen it used for comparing different models and parameters, prompts, vector-db configurations and query planning strategies. I’d love to get your feedback on it.

Would like to give a shout out to LANDrop (not affiliated) if the users are on the local network. I use it. It's very good at saturating the link bandwidth and traffic is guaranteed not to exit the firewall.

https://github.com/LANDrop/LANDrop

Related projects:

- FlyingCarpet: direct transfer over local adhoc WIFI: https://github.com/spieglt/FlyingCarpet

- LANDrop: Drop any files to any devices on your LAN: https://github.com/LANDrop/LANDrop

- In-browser file transfer similar to Airdrop: https://snapdrop.net/

- Magic Wormhole: simple file transfer from computer-to-computer over the net: https://github.com/magic-wormhole/magic-wormhole

- Croc: similar to magic wormhole: https://github.com/schollz/croc

- Wormhole: user-friendly in-browser based e2e encrypted file transfer: https://wormhole.app/

Looks like it is not. Maybe someone wants to maintain a fork.

https://github.com/LANDrop/LANDrop/issues/138

iCloud Drives doesn’t break your file structure, you just drop the whole folders to the iCloud Drives, let it sync, and move them out of iCloud Drive to the internal storage on your iPad. Alternatively there’s plenty of other apps that can do local transfer, like https://landrop.app/. Use another cloud drive service like google drives. Or setup file server on your Mac with SMB and connect it on your iPad: https://support.apple.com/en-us/guide/mac-help/mh17131/10.13/mac/10.13

i just use LANDrop. its free and has no size limitations.

=> https://landrop.app/

I've looked... but its seems there isn't an application based solution without requiring app indicator support which is something that eOS doesn't officially have. If you're comfortable modding eOS to have app indicator support you can then use LanDrop or use KDE Connect with an app indicator. You can also use sharedrop.io or snapdrop.net.