furiousteabag
Joined 5 years ago
Karma
160
Submissions
30
Member since
Nov 7
asmirnov.xyz
View on Hacker NewsRecent submissions
Comment · 1 year ago
same, did you figure it out?
Comment · 1 year ago
Another thing we are trying to understand is whether the 2D element adds value to the simulation. A simpler option would be a pure text/chat interface. Still, the hypothesis here is that it is easier to comprehend what's going on in the environment with an actual 2D world and characters, and it might be more immersive compared to just a text interface.
Comment · 2 years ago
Hey hey! We tried Clay v1 with 768 embeddings size using your tutorials. We then split NAIP SF to chips and indexed them. Afterwards, we performed image-to-image similarity search like in your explorer.
We tried to search for bridges, beaches, tennis courts, etc. It worked, but it didn't work well. The top of the ranking was filled with unrelated objects. We found that similarity scores are stacked together too much (similarity values are between 0.91 and 0.92 with 4 digit difference, ~200k tiles), so the encoder made very little difference between objects.
I believe that Clay can be used with additional fine-tuning for classification and segmentation, but standalone embeddings are pretty poor.
Check this: https://github.com/wangzhecheng/SkyScript. It is a dataset of OSM tags and satellite images. CLIP fine-tuned on that gives good embeddings for text-to-image search as well as image-to-image.
Comment · 2 years ago
Curious to see what other things you will simulate in the future!
Shameless plug: recently we've built a demo that allows you to search for objects in San Francisco using natural language. You can look for things like Tesla cars, dry patches, boats, and more. Link: https://demo.bluesight.ai/
We've tried using Clay embeddings but we quickly found out that they perform poorly for similarity search compared to embeddings produced by CLIP fine tuned on OSM captions (SkyScript).
Comment · 2 years ago
Thanks for sharing Brooklyn text demo. Haven't seen it!
Captioning images using VLM would definitely help as an additional conditional feature. Maybe it even would be enough to use only embeddings of captions to do search!
We chose aerial satellite instead of street view because we plan to apply the same technologies where street view is not available, e.g. crop fields or forests. Another thing is that we plan to monitor areas that change frequently and street view data is not enough to keep up. But the idea is great! Although your query "palace of fine arts" is not extremely exciting because it is searchable via Google Maps :D
"USF" by itself doesn't work, "USF word" pointed me where needed xD
"beach" and "picnic tables" indeed doesn't work in object mode, but works great in "big" mode, probably because they needs some context around themselves
"lots of people" didn't work, "a crowd of people" seems to work. Interesting, that almost the same (semantically) queries produce very different results!
Comment · 2 years ago
I really like using pandoc as a build system [1] for my personal website to convert .md to .html. I can use templates, automatically generate a table of content and run some lua scripts to get the desired result, such as clickable headers.
[1]: https://github.com/furiousteabag/asmirnov.xyz/blob/master/bu...
Comment · 2 years ago
You are correct, training sorely in fp16/bf16 can lead to imprecise weight updates or even gradients turning to zero. Because of that, mixed precision is used. In mixed precision training, we keep a copy of the weights in fp32 (master model) and the training loop looks like this:
compute the output with the fp16 model, then the loss
-> back-propagate the gradients in half-precision
-> copy the gradients in fp32 precision
-> do the update on the master model (in fp32 precision)
-> copy the master model in the fp16 model.
We also do loss scaling which means multiplying the output of the loss function by some scalar number before backprop (necessary in fp16 but not required in bf16).
Check out the fastai docs for more details: https://docs.fast.ai/callback.fp16.html
Comment · 2 years ago
Mixed precision is a default method to pretrain and full fine tune right now. It is especially good in transformers, because they have memory bottleneck in activations (outputs of intermediate layers stored for backprop), and running forward pass in fp16/bf16 reduces VRAM by almost half (speeds up forward pass as well).
Comment · 2 years ago
There is no option to select quantized version yet. Will work on that!