darknet

> The YOLO series is developed and maintained by Ultralytics. All YOLO code and weights are released under the AGPL-3.0 license.The YOLO series is developed and maintained by Ultralytics. All YOLO code and weights are released under the AGPL-3.0 license. The original author of YOLO and the Darknet framework [1] issued the code under pretty much every license you wish to use [2]. My preferred fork by AlexeyAB is under an equally permissive license [3]. Ultralytics then created their own model under the AGPL-3.0 license [4], which probably would never stand up in a court as they have the model from the likes of YOLOv3 in their source [5]. This entire article is flawed anyway, because they don't state which YOLOv11 model they are using or compare the accuracy. They appear to have just taken the pre-trained models and assumed it's apples-to-a

Yet another TEDIOUS BATTLE: Python vs. C++/C stack. This project gained popularity due to the HIGH DEMAND for running large models with 1B+ parameters, like `llama`. Python dominates the interface and training ecosystem, but prior to llama.cpp, non-ML professionals showed little interest in a fast C++ interface library. While existing solutions like tensorflow-serving [1] in C++ were sufficiently fast with GPU support, llama.cpp took the initiative to optimize for CPU and trim unnecessary code, essentially code-golfing and sacrificing some algorithm correctness for improved performance, which isn't favored by "ML research". NOTE: In my opinion, a true pioneer was DarkNet, which implemented the YOLO model series and significantly outperformed others [2]. Same trick basically like llama.cpp [1] https://github.com/tensorflow/serving [2] http

Yes, I have done non-trivial implementations of a number of SoTA models in Julia. The framework I've used is Flux[1] which I love for it's simplicity, it is very much like the DarkNet[2] framework in that regard which is refreshing after using TensorFlow. PyTorch is much better about not having unnecessary complexity and a sensible API but Flux is certainly better. The ability for Julia to compile directly to PTX assembly[3][4] means that you can even write the GPU kernels in Julia and eliminate the C/C++ CUDA code. Unfortunately, there is still a lot of work to be done to make it as reliably fast and easy as TensorFlow/PyTorch so I don't think it is usable for production yet. I hope it will be production ready soon but it will likely take some time to highly tune the compute stacks. They are already working on AMD GPU support with AMDGPU

I guess object detection is the bottleneck of the pipeline, which highly optimized YOLOv4-tiny[0]could get to 39 FPS on Jetson Nano. Also camera decoding with Nvidia Deepstream is a lot faster than OpenCV. [0] https://github.com/pjreddie/darknet/issues/2201

> Edit: Although as yeldarb explains in a comment here[3], > it's probably a bit more complicated than that. Legally speaking I'm not sure anything wrong was really done here. Morally speaking, it seems quite unethical. AlexeyAB has really been carrying the torch of the Darknet framework and the YOLO neural network for quite some time (with pjreddie effectively handing it over to him). AlexeyAB has been providing support on pjreddie's abandoned repository (e.g. [1]) and actively working on improvements in a fork [2]. If you look at the contributors graphs, he really has been keeping the project alive [3] (vs Darknet by pjreddie [4]). Probably the worse part in my opinion is that they have also seemingly bypassed the open source nature of the project. This is quite damning. [1] https://github.com/pjreddie/darknet/issues/1900 [2] https://gi

I really like his licensing model, specially the META license: https://github.com/pjreddie/darknet/blob/master/LICENSE.meta

My favorite (in terms of dark humor, if we’re honest) is YOLO, one of the more interesting deep learning object detectors. [1] It is the exact opposite of yours in every way. The code is brilliant however. Even the papers are snarky. [2] [1] https://github.com/pjreddie/darknet/commits/master [2] https://arxiv.org/pdf/1804.02767.pdf

If you want to take a look at a framework that isn't a terrible mess and that one person can understand both Darknet [1] and Flux [2] are really nice. Learned a lot from reading their source. [1]: https://github.com/pjreddie/darknet/ [2]: https://github.com/FluxML/Flux.jl

Paper: https://pjreddie.com/media/files/papers/YOLOv3.pdf GitHub: https://github.com/pjreddie/darknet Joseph Redmon and Ali Farhadi are funny and informative as always. Disclosure: I work for Vulcan Inc. and collaborate with AI2 regularly.

There are also tiny-dnn [1] and Darknet [2]. I'm curious how they compare to KANN. [1] https://github.com/tiny-dnn/tiny-dnn [2] https://github.com/pjreddie/darknet

The tutorial says "Edit src/yolo.c, lines 54 and 55" to train for different classes, but I don't have any yolo.c file at src folder... any ideas? [https://github.com/pjreddie/darknet/wiki/YOLO:-Real-Time-Obj...]

(Darknet) https://github.com/pjreddie/darknet