Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/apple/corenet

CoreNet: A library for training deep neural networks
https://github.com/apple/corenet

Last synced: 3 days ago
JSON representation

CoreNet: A library for training deep neural networks

Awesome Lists containing this project

README 

        
# CoreNet: A library for training deep neural networks



CoreNet is a deep neural network toolkit that allows researchers and engineers to train standard and novel small and large-scale models for variety of tasks, including foundation models (e.g., CLIP and LLM), object classification, object detection, and semantic segmentation.



## Table of contents



   * [What's new?](#whats-new)

   * [Research efforts at Apple using CoreNet](#research-efforts-at-apple-using-corenet)

   * [Installation](#installation)

   * [Directory Structure](#directory-structure)

   * [Maintainers](#maintainers)

   * [Contributing to CoreNet](#contributing-to-corenet)

   * [License](#license)

   * [Relationship with CVNets](#relationship-with-cvnets)

   * [Citation](#citation)



## What's new?



   * ***October 2024***: Version 0.1.1 of the CoreNet library includes

      * [KV Prediction](./projects/kv-prediction/)



## Research efforts at Apple using CoreNet



Below is the list of publications from Apple that uses CoreNet. Also, training and evaluation recipes, as well as links to pre-trained models, can be found inside the [projects](./projects/) folder. Please refer to it for further details.



   * [KV Prediction for Improved Time to First Token](https://arxiv.org/abs/2410.08391)

   * [OpenELM: An Efficient Language Model Family with Open Training and Inference Framework](https://arxiv.org/abs/2404.14619)

   * [CatLIP: CLIP-level Visual Recognition Accuracy with 2.7x Faster Pre-training on Web-scale Image-Text Data](https://arxiv.org/abs/2404.15653)

   * [Reinforce Data, Multiply Impact: Improved Model Accuracy and Robustness with Dataset Reinforcement](https://arxiv.org/abs/2303.08983)

   * [CLIP meets Model Zoo Experts: Pseudo-Supervision for Visual Enhancement](https://arxiv.org/abs/2310.14108)

   * [FastVit: A Fast Hybrid Vision Transformer using Structural Reparameterization](https://arxiv.org/abs/2303.14189)

   * [Bytes Are All You Need: Transformers Operating Directly on File Bytes](https://arxiv.org/abs/2306.00238)

   * [MobileOne: An Improved One millisecond Mobile Backbone](https://arxiv.org/abs/2206.04040)

   * [RangeAugment: Efficient Online Augmentation with Range Learning](https://arxiv.org/abs/2212.10553)

   * [Separable Self-attention for Mobile Vision Transformers (MobileViTv2)](https://arxiv.org/abs/2206.02680)

   * [CVNets: High performance library for Computer Vision, ACM MM'22](https://arxiv.org/abs/2206.02002)

   * [MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer, ICLR'22](https://arxiv.org/abs/2110.02178)



## Installation



You will need Git LFS (instructions below) to run tests and Jupyter notebooks 

([instructions](https://jupyter.org/install)) in this repository,

and to contribute to it so we recommend that you install and activate it first.



On Linux we recommend to use Python 3.10+ and PyTorch (version >= v2.1.0), on

macOS system Python 3.9+ should be sufficient.



Note that the optional dependencies listed below are required if you'd like to

make contributions and/or run tests.



For Linux (substitute `apt` for your package manager):



```bash

sudo apt install git-lfs



git clone [email protected]:apple/corenet.git

cd corenet

git lfs install

git lfs pull

# The following venv command is optional, but recommended. Alternatively, you can create and activate a conda environment.

python3 -m venv venv && source venv/bin/activate

python3 -m pip install --editable .

```



To install optional dependencies for audio and video processing:



```bash

sudo apt install libsox-dev ffmpeg

```



For macOS, assuming you use Homebrew:



```bash

brew install git-lfs



git clone [email protected]:apple/corenet.git

cd corenet

cd \$(pwd -P)  # See the note below.

git lfs install

git lfs pull

# The following venv command is optional, but recommended. Alternatively, you can create and activate a conda environment.

python3 -m venv venv && source venv/bin/activate

python3 -m pip install --editable .

```



To install optional dependencies for audio and video processing:



```bash

brew install sox ffmpeg

```



Note that on macOS the file system is case insensitive, and case sensitivity

can cause issues with Git. You should access the repository on disk as if the

path were case sensitive, i.e. with the same capitalization as you see when you

list the directories `ls`. You can switch to such a path with the `cd $(pwd -P)`

command.



## Directory Structure



This section provides quick access and a brief description for important CoreNet directories.



 Description 

 Quick Access 



  
 Getting Started 
 

Working with the examples is an easy way to get started with CoreNet. 

  

└── tutorials

    ├── train_a_new_model_on_a_new_dataset_from_scratch.ipynb

    ├── guide_slurm_and_multi_node_training.md

    ├── clip.ipynb

    ├── semantic_segmentation.ipynb

    └── object_detection.ipynb

  



  
 Training Recipes 


CoreNet provides reproducible training recipes, in addition to the pretrained model 

weights and checkpoints for the publications that are listed in projects/ directory.



Publication project directories generally contain the following contents:



* `README.md` provides documentation, links to the pretrained weights, and citations.

* `/.yaml` provides configuration for reproducing the trainings and evaluations.

  

└── projects

    ├── kv-prediction (*)

    ├── byteformer

    ├── catlip

    ├── clip

    ├── fastvit

    ├── mobilenet_v1

    ├── mobilenet_v2

    ├── mobilenet_v3

    ├── mobileone

    ├── mobilevit

    ├── mobilevit_v2

    ├── openelm

    ├── range_augment

    ├── resnet

    └── vit




(*) Newly released.

  



  
 MLX Examples 


MLX examples demonstrate how to run CoreNet models efficiently on Apple Silicon.

Please find further information in the README.md file within the corresponding example directory.



  

└──mlx_example

    ├── clip

    └── open_elm

  



  
 Model Implementations 
 

Models are organized by tasks (e.g. "classification"). You can find all model implementations for each

task in the corresponding task folder. 



Each model class is decorated by a 

`@MODEL_REGISTRY.register(name="", type="")` decorator. 

To use a model class in CoreNet training or evaluation,

assign `models..name = ` in the YAML configuration.



  

└── corenet

    └── modeling

        └── models

            ├── audio_classification

            ├── classification

            ├── detection

            ├── language_modeling

            ├── multi_modal_img_text

            └── segmentation

  



  
 Datasets 
 

Similarly to the models, datasets are also categorized by tasks.

  

└── corenet

    └── data

        └── datasets

            ├── audio_classification

            ├── classification

            ├── detection

            ├── language_modeling

            ├── multi_modal_img_text

            └── segmentation

  



  
 Other key directories 
 

In this section, we have highlighted the rest of the key directories that implement 

classes corresponding to the names that are referenced in the YAML configurations.

  

└── corenet

    ├── loss_fn

    ├── metrics

    ├── optims

    │   └── scheduler

    ├── train_eval_pipelines

    ├── data

    │   ├── collate_fns

    │   ├── sampler

    │   ├── text_tokenizer

    │   ├── transforms

    │   └── video_reader

    └── modeling

        ├── layers

        ├── modules

        ├── neural_augmentor

        └── text_encoders

  



## Maintainers

This code is developed and maintained by Maxwell Horton, Mohammad Sekhavat Yanzi Jin, and Dmitry Belenko.



### Previous Maintainers

* Sachin Mehta

* Farzad Abdolhosseini



## Contributing to CoreNet



We welcome PRs from the community! You can find information about contributing to CoreNet in our [contributing](CONTRIBUTING.md) document. 



Please remember to follow our [Code of Conduct](CODE_OF_CONDUCT.md).



## License



For license details, see [LICENSE](LICENSE). 



## Relationship with CVNets



CoreNet evolved from CVNets, to encompass a broader range of applications beyond computer vision. Its expansion facilitated the training of foundational models, including LLMs.



## Citation



If you find our work useful, please cite the following paper:



``` 

@inproceedings{mehta2022cvnets, 

     author = {Mehta, Sachin and Abdolhosseini, Farzad and Rastegari, Mohammad}, 

     title = {CVNets: High Performance Library for Computer Vision}, 

     year = {2022}, 

     booktitle = {Proceedings of the 30th ACM International Conference on Multimedia}, 

     series = {MM '22} 

}

```