All Projects → megvii-research → basecls

megvii-research / basecls

Licence: Apache-2.0 license
A codebase & model zoo for pretrained backbone based on MegEngine.

Programming Languages

python
139335 projects - #7 most used programming language
Makefile
30231 projects

Projects that are alternatives of or similar to basecls

Pytorch Image Models
PyTorch image models, scripts, pretrained weights -- ResNet, ResNeXT, EfficientNet, EfficientNetV2, NFNet, Vision Transformer, MixNet, MobileNet-V3/V2, RegNet, DPN, CSPNet, and more
Stars: ✭ 15,232 (+52424.14%)
Mutual labels:  pretrained-models, imagenet-classifier, distributed-training
Fengshenbang-LM
Fengshenbang-LM(封神榜大模型)是IDEA研究院认知计算与自然语言研究中心主导的大模型开源体系,成为中文AIGC和认知智能的基础设施。
Stars: ✭ 1,813 (+6151.72%)
Mutual labels:  pretrained-models, distributed-training
Zf unet 224 pretrained model
Modification of convolutional neural net "UNET" for image segmentation in Keras framework
Stars: ✭ 195 (+572.41%)
Mutual labels:  pretrained-models
Open-Source-Models
Address book for computer vision models.
Stars: ✭ 30 (+3.45%)
Mutual labels:  pretrained-models
CheXpert-Challenge
Code for CheXpert Challenge 2019 Top 1 && Top 2 solution
Stars: ✭ 30 (+3.45%)
Mutual labels:  pretrained-models
Dialogrpt
EMNLP 2020: "Dialogue Response Ranking Training with Large-Scale Human Feedback Data"
Stars: ✭ 216 (+644.83%)
Mutual labels:  pretrained-models
Awesome Pretrained Chinese Nlp Models
Awesome Pretrained Chinese NLP Models,高质量中文预训练模型集合
Stars: ✭ 195 (+572.41%)
Mutual labels:  pretrained-models
Crslab
CRSLab is an open-source toolkit for building Conversational Recommender System (CRS).
Stars: ✭ 183 (+531.03%)
Mutual labels:  pretrained-models
Learning-To-Compare-For-Text
Learning To Compare For Text , Few shot learning in text classification
Stars: ✭ 38 (+31.03%)
Mutual labels:  pretrained-models
AiSpace
AiSpace: Better practices for deep learning model development and deployment For Tensorflow 2.0
Stars: ✭ 28 (-3.45%)
Mutual labels:  pretrained-models
pinpoint-node-agent
Pinpoint Node.js agent
Stars: ✭ 45 (+55.17%)
Mutual labels:  distributed-training
pose-estimation-3d-with-stereo-camera
This demo uses a deep neural network and two generic cameras to perform 3D pose estimation.
Stars: ✭ 40 (+37.93%)
Mutual labels:  pretrained-models
Pytorch cifar10
Pretrained TorchVision models on CIFAR10 dataset (with weights)
Stars: ✭ 219 (+655.17%)
Mutual labels:  pretrained-models
vietnamese-roberta
A Robustly Optimized BERT Pretraining Approach for Vietnamese
Stars: ✭ 22 (-24.14%)
Mutual labels:  pretrained-models
horovod-ansible
Create Horovod cluster easily using Ansible
Stars: ✭ 22 (-24.14%)
Mutual labels:  distributed-training
Yolov3 Object Detection With Opencv
This project implements a real-time image and video object detection classifier using pretrained yolov3 models.
Stars: ✭ 191 (+558.62%)
Mutual labels:  pretrained-models
Keras Retinanet For Open Images Challenge 2018
Code for 15th place in Kaggle Google AI Open Images - Object Detection Track
Stars: ✭ 251 (+765.52%)
Mutual labels:  pretrained-models
object-flaw-detector-cpp
Detect various irregularities of a product as it moves along a conveyor belt.
Stars: ✭ 19 (-34.48%)
Mutual labels:  pretrained-models
masr
中文语音识别系列,读者可以借助它快速训练属于自己的中文语音识别模型,或直接使用预训练模型测试效果。
Stars: ✭ 179 (+517.24%)
Mutual labels:  pretrained-models
libai
LiBai(李白): A Toolbox for Large-Scale Distributed Parallel Training
Stars: ✭ 284 (+879.31%)
Mutual labels:  distributed-training

BaseCls

Documentation Status CI codecov

BaseCls 是一个基于 MegEngine 的预训练模型库,帮助大家挑选或训练出更适合自己科研或者业务的模型结构。

文档地址:https://basecls.readthedocs.io

安装

安装环境

BaseCls 需要 Python >= 3.6。

BaseCls 依赖 MegEngine >= 1.6.0。

通过包管理器安装

通过 pip 包管理器安装 BaseCls 的命令如下:

pip3 install basecls --user

默认不会安装包括 MegEngine 在内的部分依赖,可以通过以下命令进行完整安装:

pip3 install basecls[all] --user

对于 conda 用户, 可以选择通过在环境中先安装 pip,再按照上述方式进行 BaseCls 的安装。

通过源代码安装

为保证模型性能的可追溯性,避免实验碎片化,建议通过包管理器安装。如果包管理器安装的方式无法满足你的需求,则可以尝试自行通过源码安装。

安装依赖

pip3 install -r requirements.txt --user

安装 BaseCls

python3 setup.py develop --user

验证安装

在 Python 中导入 BaseCls 验证安装成功并查看安装版本:

import basecls
print(basecls.__version__)

开发者须知

开发环境

# 安装依赖
pip3 install -r requirements-dev.txt --user

# 配置 pre-commit
pre-commit install

开发流程

提交者需补充相应修改的单元测试。

# (外部开发者)fork repo,或(内部开发者)建立 new-feature 分支
git checkout -b new-feature

# 进行修改

# 代码风格检查与格式化
make lint
make format

# 单元测试与覆盖率检查
make unittest

# 提交修改
git commit

# 提交MR/PR
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].