models
efficientnet
Our great sponsors
models | efficientnet | |
---|---|---|
96 | 9 | |
76,598 | 2,057 | |
0.2% | - | |
9.5 | 0.0 | |
2 days ago | 3 months ago | |
Python | Python | |
GNU General Public License v3.0 or later | Apache License 2.0 |
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.
models
-
Changing box prediction head on SSD from TF2 model zoo
I am using SSD ResNet50 V1 FPN 1024x1024 (RetinaNet50) from TF model zoo .
- Labeling question
-
I'm looking for article for object detection explanation with working code
I spent some time looking for an article that explains object detection, but it seems that there are a lot of articles out there that are not very helpful. Some of these articles focus on specific things like mAP or UoI, but without the broader context, they are not very useful. The main issue with these articles is that they either don't provide any code, or they give examples that are not very helpful, like terminal commands to download a framework and train a model. I started from this link https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2.md, but it id not very useful. What I really need is a comprehensive explanation of how object detection works, along with working code that I can use to see the results for myself. I know that there are many different approaches to object localization, such as one-stage or two-stage detection, Faster R-CNN, or SSD, but I don't really care which approach will be described. I just need a starting point with clear explanations and working code that I can run.
-
good computer vision or deep learning projects in github
TensorFlow Models (GitHub: https://github.com/tensorflow/models) is a collection of diverse TensorFlow-based ML and DL models for tasks like image classification, object detection, and text classification.
- [D] I just realised: GPT-4 with image input can interpret any computer screen, any userinterface and any combination of them.
- [D]Custom Trained Networks for EasyOCR
-
Has anyone tried reverse engineering Google Tensor's AI-specific instruction set?
Assuming you're talking about leveraging the device's the device's Tensor Processing unit for machine learning then there then you're in luck because Google designed the TPU to work extremely well with the machine learning solutions developed by Google such as easy to use SDKs, robust runtimes and APIs ( e.g. - which you probably aren't going to need to touch). If you're a researcher there's plenty of lower level stuff floating about - but developers would be, again, better off staying away from it.
-
Tensorflow for M1 macs with GPU support
Thank you so that worked and I was able to install it 😅. But when I try to run the test script as mentioned here, I get an error ModuleNotFoundError: No module named 'object_detection'. Am I doing something wrong, I’m using a conda environment and I have tensorflow-macos and tensorflow-metal plug-in installed in the same environment as tf-models.
-
Object detection API deprecated
I've noticed while implementing tensorflow object detection API for a client that they have deprecated the repo and will not be updating it: https://github.com/tensorflow/models/tree/master/research/object_detection
-
NVIDIA's Rip-Off - RTX 4070 Ti Review & Benchmarks
I implore you, download a model from Tensorflow’s model repo and try training it on your conventional GPU. See how much your memory bandwidth and memory count will severely bottleneck performance, in addition see how long it takes to get any decent results.
efficientnet
-
Getting Started with Gemma Models
Examples of lightweight models include MobileNet, a computer vision model designed for mobile and embedded vision applications, EfficientDet, an object detection model, and EfficientNet, a CNN that uses compound scaling to enable better performance. All these are lightweight models from Google.
-
How did you make that?!
There was a recent paper by Facebook (2022), where they modernise a vanilla ConvNet by using the latest empirical design choices and manage to achieve state-of-the-art performance with it. This was also done before, with EffecientNet in 2019.
-
Why did the original ResNet paper not use dropout?
not true at all, plenty of sota models combines batchnorm and dropout 1. efficientnet 2. resnet rs 3. timm resnet50 (appendix)
- Increasing Model Dimensionality
- [D] How does one choose a learning rate schedule for models that take days or weeks to train?
- [D] What's the intuition behind certain CNN architectures?
-
[D] What are some interesting hidden stuff about CNNs?
Right - I think these days they do more of a balanced tradeoff between width and depth. One more recent CNN, Efficientnet, carefully chooses the width-to-depth ratio to have the best performance for a given compute budget.
-
I made an image recognition model written in NodeJs
EfficientNet a lightweight convolutional neural network architecture achieving the state-of-the-art accuracy with an order of magnitude fewer parameters and FLOPS, on both ImageNet and five other commonly used transfer learning datasets.
-
Training custom EfficientNet from scratch (greyscale)
Additionally, if you want to custom change the number of filters in the EfficientNet I would suggest using the detailed Keras implementation of the EfficientNet in this repository.
What are some alternatives?
netron - Visualizer for neural network, deep learning and machine learning models
mmpretrain - OpenMMLab Pre-training Toolbox and Benchmark
SSD-Mobilenet-Custom-Object-Detector-Model-using-Tensorflow-2 - This repository contains the script and process to create custom SSD Mobilenet model for object detection
segmentation_models - Segmentation models with pretrained backbones. Keras and TensorFlow Keras.
onnx-tensorflow - Tensorflow Backend for ONNX
label-studio - Label Studio is a multi-type data labeling and annotation tool with standardized output format
redisai-examples - RedisAI showcase
PaddleClas - A treasure chest for visual classification and recognition powered by PaddlePaddle
labelImg - LabelImg is now part of the Label Studio community. The popular image annotation tool created by Tzutalin is no longer actively being developed, but you can check out Label Studio, the open source data labeling tool for images, text, hypertext, audio, video and time-series data.
models - A collection of pre-trained, state-of-the-art models in the ONNX format
tensorboard - TensorFlow's Visualization Toolkit
image-quality-assessment - Convolutional Neural Networks to predict the aesthetic and technical quality of images.