open_clip
openpilot
Our great sponsors
open_clip | openpilot | |
---|---|---|
27 | 839 | |
8,452 | 47,461 | |
8.2% | 1.6% | |
8.2 | 10.0 | |
17 days ago | 5 days ago | |
Jupyter Notebook | Python | |
GNU General Public License v3.0 or later | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
open_clip
-
A History of CLIP Model Training Data Advances
While OpenAI’s CLIP model has garnered a lot of attention, it is far from the only game in town—and far from the best! On the OpenCLIP leaderboard, for instance, the largest and most capable CLIP model from OpenAI ranks just 41st(!) in its average zero-shot accuracy across 38 datasets.
-
How to Build a Semantic Search Engine for Emojis
Whenever I’m working on semantic search applications that connect images and text, I start with a family of models known as contrastive language image pre-training (CLIP). These models are trained on image-text pairs to generate similar vector representations or embeddings for images and their captions, and dissimilar vectors when images are paired with other text strings. There are multiple CLIP-style models, including OpenCLIP and MetaCLIP, but for simplicity we’ll focus on the original CLIP model from OpenAI. No model is perfect, and at a fundamental level there is no right way to compare images and text, but CLIP certainly provides a good starting point.
-
Database of 16,000 Artists Used to Train Midjourney AI Goes Viral
It is a misconception that Adobe's models have not been trained on copyrighted work. Nobody should be repeating their marketing claims.
Adobe has not shown how they train the text encoders in Firefly, or what images were used for the text-based conditioning (i.e. "text to image") part of their image generation model. They are almost certainly using CLIP or T5, which are trained on LAION2b, an image dataset with the very problems they are trying to address, C4 (a text dataset similarly encumbered) and similar.
I welcome anyone who works at Adobe to simply answer this question of how they trained the text encoders for text conditioning and put it to rest. There is absolutely nothing sensitive about the issue, unless it exposes them in a lie.
So no chance. I think it's a big fat lie. They'd have to have made some other scientific breakthrough, which they didn't.
Using information from https://openai.com/research/clip and https://github.com/mlfoundations/open_clip, it's possible to investigate the likelihood that using just their stock image dataset, can they make a working text encoder?
It's certainly not impossible, but it's impracticable. On 248m images (roughly the size of Adobe Stock), CLIP gets 37% on ImageNet, and on the 2000m from LAION, it performs 71-80%. And even with 2000m images, CLIP is substantially worse performing than the approach that Imagen uses for "text comprehension," which relies on essentially many billions more images and text tokens.
-
MetaCLIP – Meta AI Research
https://github.com/mlfoundations/open_clip/blob/main/docs/op...
-
COMFYUI SDXL WORKFLOW INBOUND! Q&A NOW OPEN! (WIP EARLY ACCESS WORKFLOW INCLUDED!)
in the modal card it says: pretrained text encoders (OpenCLIP-ViT/G and CLIP-ViT/L).
-
Is Nicholas Renotte a good guide for a person who knows nothing about ML?
also, if you describe your task a bit more, we might be able to direct you to a fairly out-of-the-box solution, e.g. you might be able to use one of the pretrained models supported by https://github.com/mlfoundations/open_clip without any additional training
-
Generate Image from Vector Embedding
It says on the Stable Diffusion Github repo that it uses the “OpenCLIP-ViT/H” https://github.com/mlfoundations/open_clip model as a text encoder, and from my prior experience with CLIP, I have found that it is very easy to generate image and text embeddings (because CLIP is a multimodal model).
-
What's up in the Python community? – April 2023
https://replicate.com/pharmapsychotic/clip-interrogator
using:
cfg.apply_low_vram_defaults()
interrogate_fast()
I tried lighter models like vit32/laion400 and others etc all are very very slow to load or use (model list: https://github.com/mlfoundations/open_clip)
I'm desperately looking for something more modest and light.
-
Low accuracy on my CNN model.
A library that is very useful for this kind of application is timm. You may also find the feature representation provided by a CLIP model particularly powerful.
- Looking for OpenAI CLIP alternative
openpilot
-
Tinygrad: Hacked 4090 driver to enable P2P
Yes, but he spent several years in self-driving cars (https://comma.ai), which while interesting is also a space that a lot of players are in, so it's not the same as seeing him back to doing stuff that's a little more out there, especially as pertains to IP.
-
Imitation Learning
We have a product for sale: https://comma.ai
We raised $18.1M and have made $28M in lifetime revenue to date.
Where are you getting your narrative?
-
Driverless cars immune from traffic tickets in California under current laws
What about comma? https://comma.ai/ Seems like our old friend geohot built exactly what you want.
Positive HN discussion: https://news.ycombinator.com/item?id=36927971
-
No USS?
The issue was that the front camera on the windshield couldn’t see under the hood. You misunderstand how easy it is to solve for depth and distance with AI without requiring stereo cameras. Read https://github.com/commaai/openpilot
-
What car should I get for Seattle city and some ski/hike driving? Or not get a car at all?
Nice to have: I want to get a self-driving add-on that supports some cars better than others. Not a must but high up on my nice-to-have list.
- I need some help understanding video uploads.
-
I am nearing the end of my Kona 2020 lease, and I have an appointment at a dealer tomorrow had some questions about leasing an ioniq 6, hopefully someone can help me out.
EDIT: I probably should have added that I currently have the base model of the Kona the lowest model available, and I am looking for a similar thing in the ioniq 6, because my understanding is that it's fully compatible with the comma.ai device and therefore I am not planning on getting the better on board driving system, the Kona that I got unfortunately was not compatible with that device.
-
Tesla: Security Vulnerabilities
I wonder how bad this is compared to the competition. https://comma.ai allows you to add self-driving features to a large number of non-Tesla cars so, if we’re including physical firmware hacks as a threat vector, I’d bet tons of alternative cars (new enough Honda Odysseys, Toyota Siennas, etc: probably anything with adaptive cruise control and lane following) have the same sort of potential vulnerability.
- 2024 highlander has Toyota Security Key Now
-
Cruise co-founder and CEO Kyle Vogt resigns
Not sure, but from the first article from 4 years ago:
>Last month, we had 1,209 cars drive a little over 1,000,000 miles
Let's say they've had zero growth since then, so 48,000,000 conservatively?
Actually, from their website [1]:
>100+ million miles driven and 10k users.
[1]: https://comma.ai
What are some alternatives?
CLIP - CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
sunnypilot - sunnypilot is a fork of comma.ai's openpilot, an open source driver assistance system. sunnypilot offers the user a unique driving experience for over 290 supported car makes and models with modified behaviors of driving assist engagements. sunnypilot complies with comma.ai's safety rules as accurately as possible.
DALLE-pytorch - Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch
opendbc - democratize access to car decoder rings
taming-transformers - Taming Transformers for High-Resolution Image Synthesis
carla - Open-source simulator for autonomous driving research.
Dreambooth-Stable-Diffusion - Implementation of Dreambooth (https://arxiv.org/abs/2208.12242) with Stable Diffusion
dragonpilot - dragonpilot - 基於 openpilot 的開源駕駛輔助系統
bitsandbytes - Accessible large language models via k-bit quantization for PyTorch.
label-studio - Label Studio is a multi-type data labeling and annotation tool with standardized output format
clip-retrieval - Easily compute clip embeddings and build a clip retrieval system with them
netron - Visualizer for neural network, deep learning and machine learning models