mmdetection
pytorch-lightning
Our great sponsors
mmdetection | pytorch-lightning | |
---|---|---|
23 | 19 | |
27,742 | 19,188 | |
2.3% | - | |
8.7 | 9.9 | |
10 days ago | almost 2 years ago | |
Python | Python | |
Apache License 2.0 | 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.
mmdetection
-
Semantic segementation
When I look for benchmarks I always start here https://paperswithcode.com/task/instance-segmentation/codeless it has the lists of datasets to measure models accross lots o papers. Many are very specific models with low support or community but it gives you a good idea of the state of the art. It also lists repositories related to good community. https://github.com/open-mmlab/mmdetection seems very active and the one that is being used the most, you could use the models that it has integrated in its model zoo, within the same repository. It has the benchmarks to compare those same models and some of them are from 2022
-
How to Convert Model Mask into Polygon and save JSON?
MODEL: https://github.com/open-mmlab/mmdetection
-
Object Detection Model for Custom Dataset Training?
Would it make sense to work with OpenMMLab (https://github.com/open-mmlab/mmdetection) or Pytorch-image-models (https://github.com/rwightman/pytorch-image-models#models) since they offer a variety of models?
-
[P] Image search with localization and open-vocabulary reranking.
I wanted to have a few choices getting localization into image search (index and search time). I immediately thought of using a region proposal network (rpn) from mask-rcnn to create patches that can also be indexed and searched (and add the localisation). I figured it might be somewhat agnostic to classes. I did not want to use mmdetection or detectron2 due to their dependencies and just getting the rpn was not worth it. I was encouraged by the PyTorch native implementations of detection/segmentation models but ended up finding yolox the best.
-
MMDeploy: Deploy All the Algorithms of OpenMMLab
MMDetection: OpenMMLab detection toolbox and benchmark.
-
Removing the bounding box generated by OnnxRuntime segmentation
I have a semantic segmentation model trained using the mmdetection repo. Then it is converted to the ONNX format using the mmdeploy repo.
-
Keras vs Tensorflow vs Pytorch for a Final year Project
E.g. If you consider it an object detection problem it is: detect and localise all the pedestrians in a frame, and classify them by their (intended) action. IMO the easiest way to do this would be with mmdetection, which is built on top of pytorch. Just label your dataset, build a config, and boom you have a model. Inference with that model in only a few lines of code, you won't really need to learn too much to get started.
-
DeepSort with PyTorch(support yolo series)
MMDetection
-
[D] Pre-trained networks and batch normalization
For example, in mmdetection, they expose options in their config & implementation to freeze batch norm layers in backbones and in this config, norm_eval is set to True meaning to freeze tracking of batch norm stats, while the ResNet backbone is frozen up to the 1st stage. Example of their backbone implementation can be found here.
-
Config files in plain Python
MMDetection uses config Python scripting. It's easier to define nn.Module objects other than writing class name in a json config file
pytorch-lightning
-
Problem with pytorch lightning and optuna with multiple callbacks
def on_validation_end(self, trainer: Trainer, pl_module: LightningModule) -> None: # Trainer calls `on_validation_end` for sanity check. Therefore, it is necessary to avoid # calling `trial.report` multiple times at epoch 0. For more details, see # https://github.com/PyTorchLightning/pytorch-lightning/issues/1391. if trainer.sanity_checking: return
-
Please comment on my planned research project structure
Under the hood, the ModelWrapper object will create a ML model based on the config (so far, an XGBoost model and a PyTorch Lightning model). Each of those will have a wrapper that conducts training and evaluation (since from my understanding of Lightning, Trainers are required to be outside of the class). In lack of a better name, I call these wrappers Fitters. For uniformity, I thought about adding a common interface IFitter, which is inherited by all model wrappers as outlined below.
-
Watch out for the (PyTorch) Lightning
Join their Slack to ask the community questions and check out the GitHub here.
-
[P] Composer: a new PyTorch library to train models ~2-4x faster with better algorithms
Pytorch lightning benchmarks against pytorch on every PR (benchmarks to make sure that it is mot slower.
-
[D] What Repetitive Tasks Related to Machine Learning do You Hate Doing?
There is already a ton of momentum around automating ML workflows. I would suggest you contribute to a preexisting project like, for instance, PyTorch Lightning or fast.ai.
- PyTorch Lightening
-
[D] Are you using PyTorch or TensorFlow going into 2022?
Is the problem the sheer number of options, or the fact that they are all together in one place? Would it be better if they were organized into the different trainer entrypoints (fit, validate, ...)? If that is the case, there was an RFC proposing this which you might find interesting, feel free to drop by and comment on the issue: https://github.com/PyTorchLightning/pytorch-lightning/issues/10444
-
[D] Colab TPU low performance
I wanted to make a quick performance comparison between the GPU (Tesla K80) and TPU (v2-8) available in Google Colab with PyTorch. To do so quickly, I used an MNIST example from pytorch-lightning that trains a simple CNN.
-
[D] How to avoid CPU bottlenecking in PyTorch - training slowed by augmentations and data loading?
We've noticed GPU 0 on our 3 GPU system is sometimes idle (which would explain performance differences). However its unclear to us why that may be. Similar to this issue
-
[P] An introduction to PyKale https://github.com/pykale/pykale, a PyTorch library that provides a unified pipeline-based API for knowledge-aware multimodal learning and transfer learning on graphs, images, texts, and videos to accelerate interdisciplinary research. Welcome feedback/contribution!
If you want a good example for reference, take a look at Pytorch Lightning's readme (https://github.com/PyTorchLightning/pytorch-lightning) It answers the 3 questions of "what is this", "why should I care", and "how do i use it" almost instantly
What are some alternatives?
detectron2 - Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.
pytorch-grad-cam - Advanced AI Explainability for computer vision. Support for CNNs, Vision Transformers, Classification, Object detection, Segmentation, Image similarity and more.
yolov5 - YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite
PaddleDetection - Object Detection toolkit based on PaddlePaddle. It supports object detection, instance segmentation, multiple object tracking and real-time multi-person keypoint detection.
fastai - The fastai deep learning library
mmdetection3d - OpenMMLab's next-generation platform for general 3D object detection.
composer - Supercharge Your Model Training
sahi - Framework agnostic sliced/tiled inference + interactive ui + error analysis plots
sparktorch - Train and run Pytorch models on Apache Spark.
Mask_RCNN - Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow
Keras - Deep Learning for humans