From f9478c626cddfe81c1b078c5884be78d1bb486b4 Mon Sep 17 00:00:00 2001 From: Genius Patrick <74176172+geniuspatrick@users.noreply.github.com> Date: Wed, 17 Jan 2024 17:23:13 +0800 Subject: [PATCH] bump version to 0.3.0 (#751) --- README.md | 48 +++++++++++++++++++++++++++++++---------------- README_CN.md | 31 ++++++++++++++++++++++++++++++ RELEASE.md | 33 ++++++++++++++++++++++++++++++++ mindcv/version.py | 2 +- 4 files changed, 97 insertions(+), 17 deletions(-) diff --git a/README.md b/README.md index 346cd0bff..fa269e97e 100644 --- a/README.md +++ b/README.md @@ -295,22 +295,38 @@ Please see [configs](./configs) for the details about model performance and pret ## What is New -- 2023/6/16 -1. New version `0.2.2` is released! We upgrade to support `MindSpore` v2.0 while maintaining compatibility of v1.8 -2. New models: - - [ConvNextV2](configs/convnextv2) - - mini of [CoAT](configs/coat) - - 1.3 of [MnasNet](configs/mnasnet) - - AMP(O3) version of [ShuffleNetV2](configs/shufflenetv2) -3. New features: - - Gradient Accumulation - - DynamicLossScale for customized [TrainStep](mindcv/utils/train_step.py) - - OneCycleLR and CyclicLR learning rate scheduler - - Refactored Logging - - Pyramid Feature Extraction -4. Bug fixes: - - Serving Deployment Tutorial(mobilenet_v3 doesn't work on ms1.8 when using Ascend backend) - - Some broken links on our documentation website. +- 2024/1/17 + +Release `0.3.0` is published. We will drop MindSpore 1.x in the future release. + +1. New models: + - Y-16GF of [RegNet](configs/regnet) + - [SwinTransformerV2](configs/swintransformerv2) + - [VOLO](configs/volo) + - [CMT](configs/cmt) + - [HaloNet](configs/halonet) + - [SSD](examples/det/ssd) + - [DeepLabV3](examples/seg/deeplabv3) + - [CLIP](examples/clip) & [OpenCLIP](examples/open_clip) +2. Features: + - AsymmetricLoss & JSDCrossEntropy + - Augmentations Split + - Customized AMP +3. Bug fixes: + - Since the classifier weights are not fully deleted, you may encounter an error passing in the `num_classes` when creating a pre-trained model. +4. Refactoring: + - The names of many models have been refactored for better understanding. + - [Script](mindcv/models/vit.py) of `VisionTransformer`. + - [Script](train_with_func.py) of Mixed(PyNative+jit) mode training. +5. Documentation: + - A guide of how to extract multiscale features from backbone. + - A guide of how to finetune the pre-trained model on a custom dataset. +6. BREAKING CHANGES: + - We are going to drop support of MindSpore 1.x for it's EOL. + - Configuration `filter_bias_and_bn` will be removed and renamed as `weight_decay_filter`, + due to a prolonged misunderstanding of the MindSpore optimizer. + We will migrate the existing training recipes, but the signature change of function `create_optimizer` will be incompatible + and the old version training recipes will also be incompatible. See [PR/752](https://github.com/mindspore-lab/mindcv/pull/752) for details. See [RELEASE](RELEASE.md) for detailed history. diff --git a/README_CN.md b/README_CN.md index e02a5cd53..7977d4987 100644 --- a/README_CN.md +++ b/README_CN.md @@ -298,6 +298,37 @@ python train.py --model=resnet50 --dataset=cifar10 \ ## 更新 +- 2024/1/17 + +新版本`0.3.0`发布。我们将在未来的发布中丢弃对MindSpore1.x版本的支持 + +1. 新模型: + - [RegNet](configs/regnet)的Y-16GF规格 + - [SwinTransformerV2](configs/swintransformerv2) + - [VOLO](configs/volo) + - [CMT](configs/cmt) + - [HaloNet](configs/halonet) + - [SSD](examples/det/ssd) + - [DeepLabV3](examples/seg/deeplabv3) + - [CLIP](examples/clip) & [OpenCLIP](examples/open_clip) +2. 特性: + - 损失函数AsymmetricLoss及JSDCrossEntropy + - 数据增强分离(Augmentations Split) + - 自定义混合精度策略 +3. 错误修复: + - 由于分类器参数未完全弹出，您在新建预训练模型时传入参数`num_classes`可能会导致错误。 +4. 重构: + - 许多模型的名字发生了变更，以便更好的理解。 + - `VisionTransformer`的模型定义[脚本](mindcv/models/vit.py)。 + - 混合模式(PyNative+jit)的训练[脚本](train_with_func.py)。 +5. 文档: + - 如何提取多尺度特征的教程指引。 + - 如何在自定义数据集上微调预训练模型的教程指引。 +6. BREAKING CHANGES: + - 我们将在此小版本的未来发布中丢弃对MindSpore1.x的支持。 + - 配置项`filter_bias_and_bn`将被移除并更名为`weight_decay_filter`。 + 我们会对已有训练策略进行迁移，但函数`create_optimizer`的签名变更将是不兼容的，且未迁移旧版本的训练策略也将变得不兼容。详见[PR/752](https://github.com/mindspore-lab/mindcv/pull/752)。 + - 2023/6/16 1. 新版本 `0.2.2` 发布啦！我们将`MindSpore`升级到了2.0版本，同时保持了对1.8版本的兼容 2. 新模型: diff --git a/RELEASE.md b/RELEASE.md index b14f07a30..a1b3c72f6 100644 --- a/RELEASE.md +++ b/RELEASE.md @@ -1,5 +1,38 @@ # Release Note +## 0.3.0 (2024/1/17) + +Release `0.3.0` is published. We will drop MindSpore 1.x in the future release. + +1. New models: + - Y-16GF of [RegNet](configs/regnet) + - [SwinTransformerV2](configs/swintransformerv2) + - [VOLO](configs/volo) + - [CMT](configs/cmt) + - [HaloNet](configs/halonet) + - [SSD](examples/det/ssd) + - [DeepLabV3](examples/seg/deeplabv3) + - [CLIP](examples/clip) & [OpenCLIP](examples/open_clip) +2. Features: + - AsymmetricLoss & JSDCrossEntropy + - Augmentations Split + - Customized AMP +3. Bug fixes: + - Since the classifier weights are not fully deleted, you may encounter an error passing in the `num_classes` when creating a pre-trained model. +4. Refactoring: + - The names of many models have been refactored for better understanding. + - [Script](mindcv/models/vit.py) of `VisionTransformer`. + - [Script](train_with_func.py) of Mixed(PyNative+jit) mode training. +5. Documentation: + - A guide of how to extract multiscale features from backbone. + - A guide of how to finetune the pre-trained model on a custom dataset. +6. BREAKING CHANGES: + - We are going to drop support of MindSpore 1.x for it's EOL. + - Configuration `filter_bias_and_bn` will be removed and renamed as `weight_decay_filter`, + due to a prolonged misunderstanding of the MindSpore optimizer. + We will migrate the existing training recipes, but the signature change of function a will be incompatible + and the old version training recipes will also be incompatible. See [PR/752](https://github.com/mindspore-lab/mindcv/pull/752) for details. + ## 0.2.2 (2023/6/16) 1. New version `0.2.2` is released! We upgrade to support `MindSpore` v2.0 while maintaining compatibility of v1.8 diff --git a/mindcv/version.py b/mindcv/version.py index f2c68c62b..e5ded93b9 100644 --- a/mindcv/version.py +++ b/mindcv/version.py @@ -1,2 +1,2 @@ """version init""" -__version__ = "0.2.2" +__version__ = "0.3.0"