yeldarb

9 ms·

61.
by yeldarb 3y ago
Yeah, inference[1] is our open source package for running locally (either directly in Python or via a Docker container). It works with all the models on Universe, models you train yourself (assuming we support the architecture; we have a bu
62.
by yeldarb 3y ago
Nice, makes sense to chain models so you don't waste the attention of the big smart model on the grunt work of JSON structure. I bet there are some good post-processing heuristics you could also apply for hallucination with a flag for
63.
by yeldarb 3y ago
Will they release the code/weights?
64.
Roboflow (YC S20) is hiring a new leader for its machine learning team(ycombinator.com)
1 points by yeldarb 3y ago
65.
by yeldarb 3y ago
How can a project with 260 contributors and no CLA change their license? Wouldn’t they have to get approval from all of those contributors or remove all that code?
66.
Roboflow (YC S20) is hiring an engineer focused on their open source projects(roboflow.com)
1 points by yeldarb 3y ago
67.
by yeldarb 3y ago
In the simplest case, { “a”: 1, “b”: 2 } seems equally as valid as { “b”: 2, “a”: 1 } (also other things like different key names and white space seem like they shouldn’t effect the “rightness”.
68.
by yeldarb 3y ago
How are you measuring “rightness”? Does it have to exactly match the predefined JSON output? Or is there some qualitative “sameness” check going on? If you’re not already, you could probably give GPT-4 the correct JSON and the predicted JSO
69.
by yeldarb 3y ago
Cool to read. Looks very familiar. I started an Easter Egg Hunt Facebook game out of my dorm room in 2008 called Hatchlings. We actually had them beat in the engineers to users ratio. When we had 5 million users and more pageviews/mo t
70.
by yeldarb 3y ago
OpenAI calls it GPT-V https://help.openai.com/en/articles/8555496-gpt-v-api
71.
by yeldarb 3y ago
Weird that the object detection prompt refused to answer. What other variations of that have we tried? Do you think that's an intentional task that was RLHF'd out or a quirk of some kind? Thinking out loud some things I'd try
72.
by yeldarb 3y ago
We put the image portion through its paces and compared it with GPT-V here: https://blog.roboflow.com/first-impressions-with-google-gemi...
73.
by yeldarb 3y ago
We’ve been seeing this in vision too. I’m really excited about multimodal LLMs but it seems like they’re going to be most useful for solving new types of (qualitative) problems vs taking over (quantitative) problems where small, fast CNNs h
74.
by yeldarb 3y ago
Yeah, check out LLaVA-Plus (they call the experts in your vocabulary "tools") https://github.com/LLaVA-VL/LLaVA-Plus-Codebase
75.
by yeldarb 3y ago
I'd love one of these on my wall. Imagining a framed version of the Einstein pop-art one where the circle in the middle rotates (either periodically or via a manual lever).
76.
by yeldarb 3y ago
We tried Inferentia & though they claim their compiler should "just work" on your existing models it was producing a bunch of NaNs for us & was opaque enough that we didn't really have a path forward so we gave up on
77.
by yeldarb 3y ago
Which part of copyright law are model weights governed by? (Or, if not by copyright law, what's the legal basis that would let you choose a "license" for model weights?)
78.
by yeldarb 3y ago
Looks like the Vicuna repo is Apache 2.0 also[1]. What's the interpretation of copyright law that would prevent the code being Apache 2.0 based on the source of the fine-tuning dataset? [1] https://github.com/lm-sys&#x2
79.
by yeldarb 3y ago
This looks like a really cool project; would you be open to us PR'ing support for the 50k fine-tuned models on Roboflow Universe[1] via an `inference`[2] integration? [1] https://roboflow.com/universe [2] https:/
80.
by yeldarb 3y ago
Pretty neat! We've been using Lambda for ML serving low-volume CV models (and my understanding is AWS' Sagemaker Serverless is a lambda wrapper) for a couple of years at Roboflow and it is really good for low-volume and bursty use
81.
by yeldarb 3y ago
Are there situations where you'd want to use something like this over a more modern algorithm like ByteTrack or DeepSORT?
82.
What's inside Apple's $129 Thunderbolt cable?(twitter.com)
3 points by yeldarb 3y ago|3 comments
83.
by yeldarb 3y ago
This looks epic. Definitely going to explore adding it to Autodistill[1] this weekend. Any chance you'll be publicly releasing the internal OCR finetune? [1] https://github.com/autodistill/autodistill
84.
by yeldarb 3y ago
Looks like the (Ghost?) image CDN got hugged to death. We'll update the URLs. ``` 403. That’s an error. Your client does not have permission to get URL ... from this server. (Client IP address: ...) Rate-limit exceeded That’s all we kn
85.
by yeldarb 3y ago
If true, this is wild. I suppose a human could spend 10 seconds per Captcha, so they could do 360 per hour. Add some overhead for not being operating at peak performance every minute of every hour & call it 250. Let's say you can h
86.
by yeldarb 3y ago
I’m intrigued to see what kind of problems it’s going to be good/bad at. I think it’s going to be tricky to evaluate though because it has probably memorized all the easy images to eval it with. Eg anything pulled from Google Images (l
87.
How not to be fooled by viral charts(noahpinion.blog)
2 points by yeldarb 3y ago|0 comments
88.
by yeldarb 3y ago
Yes, GroundingDINO is an open set object detector. There are some others (eg DETIC and OWL-ViT) as well. We’ve been working on using them (often in conjunction with SAM) for auto-labeling datasets to train smaller faster models that can run
89.
by yeldarb 3y ago
Is $200/mo really a huge barrier? That's about the equivalent of 1 college credit at a local University & you'll probably learn a ton more.
90.
by yeldarb 3y ago
Yes, we use https://www.pickleai.com and it's great. You don't have to sit in meetings "just in case"; people can just loop you in afterwards and you can skim the transcript or watch on 2x speed.

More ›