yolov5js VS notebooks

I love this! TF.js is big! And to answer your question - sure, we can run that model in the browser. This is the YOLOv5 model it can be converted from PyTorch to TF.js with this script: https://github.com/ultralytics/yolov5/blob/master/export.py And then run it with my NPM package https://github.com/SkalskiP/yolov5js. ML in Java Script is the future! The problem is I don't know anything about any good tracker implemented in JS.

makesense.ia is certainly the largest one. I just recently started https://github.com/SkalskiP/yolov5js with the aim to make it much easier for frontend developers without computer vision background to use object detection in their projects. Apart from that, I have https://github.com/SkalskiP/ILearnDeepLearning.py which is a repository containing examples related to my blog posts on Medium https://medium.com/@piotr.skalski92.

By the way, I have created an NPM package, which can also make it easier for you to deploy YOLOv5 in the browser. https://github.com/SkalskiP/yolov5js

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 bunch of notebooks available[2]), or train in our platform, plus several more general foundation models[3] (for things like embeddings, zero-shot detection, question answering, OCR, etc).

We also have a hosted API[4] you can hit for most models we support (except some of the large vision models that are really GPU-heavy) if you prefer.

[1] https://github.com/roboflow/inference

[2] https://github.com/roboflow/notebooks

[3] https://inference.roboflow.com/foundation/about/

[4] https://docs.roboflow.com/deploy/hosted-api

We (the Roboflow open source team) actively write open source Google Colab notebooks showing how to use new SOTA models. Our library covers SAM, CLIP, Detectron2, YOLOv8, RTMDet, DINOv2, and more. These notebooks helped me cross the chasm from "how do I use X model?" to being able to both write and understand inference code.

Here's an open source SegFormer notebook and guide: https://github.com/roboflow/notebooks/blob/main/notebooks/train-segformer-segmentation-on-custom-data.ipynb

Great ways to get some experience in general ML: * https://kaggle.com/learn to up your skill-set, practice a bit, and improve breadth of knowledge in topics like deep learning and computer vision * https://huggingface.co/learn free NLP courses that will really beef up your skillset * https://madewithml.com - robust tutorials for the end-to-end deep learning MLOps process * https://roboflow.com/learn - intro course material and some advanced topics in computer vision; tutorial walkthroughs for model training: https://github.com/roboflow/notebooks

Repo: https://github.com/roboflow/notebooks/blob/main/notebooks/sa...

Here are the notebooks I spend day and night refining: https://github.com/roboflow/notebooks

You’ll just need to have labeled solar panel images, and pick a model architecture and tutorial to train with: https://github.com/roboflow/notebooks

dataset: https://universe.roboflow.com/roboflow-jvuqo/football-players-detection-3zvbc/dataset/4 code: https://github.com/roboflow/notebooks/blob/main/notebooks/how-to-track-football-players.ipynb video: https://youtu.be/QCG8QMhga9k

Have fun! Great field. Just also try out the first few OpenCV tutorials, and train a few custom model to deploy to see what you think. Here’s a ton of free open source notebooks: https://github.com/roboflow/notebooks