Skip to content

Commit

Permalink
Merge 91f9bfa into 194c1c9
Browse files Browse the repository at this point in the history
  • Loading branch information
@xvjiarui
xvjiarui authored Jun 27, 2021
2 parents 194c1c9 + 91f9bfa commit 1105422
Showing 1 changed file with 44 additions and 55 deletions.
99 changes: 44 additions & 55 deletions docs_zh-CN/model_zoo.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -2,156 +2,145 @@

## 共同设定

* We use distributed training with 4 GPUs by default.
* All pytorch-style pretrained backbones on ImageNet are train by ourselves, with the same procedure in the [paper](https://arxiv.org/pdf/1812.01187.pdf).
Our ResNet style backbone are based on ResNetV1c variant, where the 7x7 conv in the input stem is replaced with three 3x3 convs.
* For the consistency across different hardwares, we report the GPU memory as the maximum value of `torch.cuda.max_memory_allocated()` for all 4 GPUs with `torch.backends.cudnn.benchmark=False`.
Note that this value is usually less than what `nvidia-smi` shows.
* We report the inference time as the total time of network forwarding and post-processing, excluding the data loading time.
Results are obtained with the script `tools/benchmark.py` which computes the average time on 200 images with `torch.backends.cudnn.benchmark=False`.
* There are two inference modes in this framework.

* `slide` mode: The `test_cfg` will be like `dict(mode='slide', crop_size=(769, 769), stride=(513, 513))`.

In this mode, multiple patches will be cropped from input image, passed into network individually.
The crop size and stride between patches are specified by `crop_size` and `stride`.
The overlapping area will be merged by average

* `whole` mode: The `test_cfg` will be like `dict(mode='whole')`.

In this mode, the whole imaged will be passed into network directly.

By default, we use `slide` inference for 769x769 trained model, `whole` inference for the rest.
* For input size of 8x+1 (e.g. 769), `align_corner=True` is adopted as a traditional practice.
Otherwise, for input size of 8x (e.g. 512, 1024), `align_corner=False` is adopted.
* 我们默认使用 4 卡分布式训练
* 所有 PyTorch 风格的 ImageNet 预训练网络由我们自己训练，和 [论文](https://arxiv.org/pdf/1812.01187.pdf) 保持一致。
我们的 ResNet 网络是基于 ResNetV1c 的变种，在这里输入层的 7x7 卷积被 3个 3x3 取代。
* 为了在不同的硬件上保持一致，我们以 `torch.cuda.max_memory_allocated()` 的最大值作为 GPU 占用率，同时设置 `torch.backends.cudnn.benchmark=False`。
注意，这通常比 `nvidia-smi` 显示的要少。
* 我们以网络 forward 和后处理的时间加和作为推理时间，除去数据加载时间。我们使用脚本 `tools/benchmark.py` 来获取推理时间，它在 `torch.backends.cudnn.benchmark=False` 的设定下，计算 200 张图片的平均推理时间。
* 在框架中，有两种推理模式。
* `slide` 模式（滑动模式）：测试的配置文件字段 `test_cfg` 会是 `dict(mode='slide', crop_size=(769, 769), stride=(513, 513))`.
在这个模式下，从原图中裁剪多个小图分别输入网络中进行推理。小图的大小和小图之间的距离由 `crop_size` 和 `stride` 决定，重合区域会进行平均。
* `whole` 模式 （全图模式）:测试的配置文件字段 `test_cfg` 会是 `dict(mode='whole')`. 在这个模式下，全图会被直接输入到网络中进行推理。
对于 769x769 下训练的模型，我们默认使用 `slide` 进行推理，其余模型用 `whole` 进行推理。
* 对于输入大小为 8x+1 （比如769），我们使用 `align_corners=True`。其余情况，对于输入大小为 8x+1 (比如 512，1024)，我们使用 `align_corners=False`。

## 基线

### FCN

Please refer to [FCN](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/fcn) for details.
请参考 [FCN](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/fcn) for details.

### PSPNet

Please refer to [PSPNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet) for details.
请参考 [PSPNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet) for details.

### DeepLabV3

Please refer to [DeepLabV3](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3) for details.
请参考 [DeepLabV3](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3) for details.

### PSANet

Please refer to [PSANet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/psanet) for details.
请参考 [PSANet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/psanet) for details.

### DeepLabV3+

Please refer to [DeepLabV3+](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3plus) for details.
请参考 [DeepLabV3+](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3plus) for details.

### UPerNet

Please refer to [UPerNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/upernet) for details.
请参考 [UPerNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/upernet) for details.

### NonLocal Net

Please refer to [NonLocal Net](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/nlnet) for details.
请参考 [NonLocal Net](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/nlnet) for details.

### EncNet

Please refer to [EncNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/encnet) for details.
请参考 [EncNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/encnet) for details.

### CCNet

Please refer to [CCNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/ccnet) for details.
请参考 [CCNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/ccnet) for details.

### DANet

Please refer to [DANet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/danet) for details.
请参考 [DANet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/danet) for details.

### APCNet

Please refer to [APCNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/apcnet) for details.
请参考 [APCNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/apcnet) for details.

### HRNet

Please refer to [HRNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/hrnet) for details.
请参考 [HRNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/hrnet) for details.

### GCNet

Please refer to [GCNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/gcnet) for details.
请参考 [GCNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/gcnet) for details.

### DMNet

Please refer to [DMNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/dmnet) for details.
请参考 [DMNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/dmnet) for details.

### ANN

Please refer to [ANN](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/ann) for details.
请参考 [ANN](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/ann) for details.

### OCRNet

Please refer to [OCRNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/ocrnet) for details.
请参考 [OCRNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/ocrnet) for details.

### Fast-SCNN

Please refer to [Fast-SCNN](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/fastscnn) for details.
请参考 [Fast-SCNN](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/fastscnn) for details.

### ResNeSt

Please refer to [ResNeSt](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/resnest) for details.
请参考 [ResNeSt](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/resnest) for details.

### Semantic FPN

Please refer to [Semantic FPN](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/semfpn) for details.
请参考 [Semantic FPN](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/semfpn) for details.

### PointRend

Please refer to [PointRend](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/point_rend) for details.
请参考 [PointRend](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/point_rend) for details.

### MobileNetV2

Please refer to [MobileNetV2](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/mobilenet_v2) for details.
请参考 [MobileNetV2](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/mobilenet_v2) for details.

### MobileNetV3

Please refer to [MobileNetV3](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/mobilenet_v3) for details.
请参考 [MobileNetV3](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/mobilenet_v3) for details.

### EMANet

Please refer to [EMANet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/emanet) for details.
请参考 [EMANet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/emanet) for details.

### DNLNet

Please refer to [DNLNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/dnlnet) for details.
请参考 [DNLNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/dnlnet) for details.

### CGNet

Please refer to [CGNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/cgnet) for details.
请参考 [CGNet](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/cgnet) for details.

### Mixed Precision (FP16) Training

Please refer [Mixed Precision (FP16) Training](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/fp16/README.md) for details.

## Speed benchmark
## 速度标定

### Hardware
### 硬件

* 8 NVIDIA Tesla V100 (32G) GPUs
* Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz

### Software environment
### 软件环境

* Python 3.7
* PyTorch 1.5
* CUDA 10.1
* CUDNN 7.6.03
* NCCL 2.4.08

### Training speed
### 训练速度

For fair comparison, we benchmark all implementations with ResNet-101V1c.
The input size is fixed to 1024x512 with batch size 2.
为了公平比较，我们全部使用 ResNet-101V1c 进行标定。输入大小为 1024x512，批量样本数为 2。

The training speed is reported as followed, in terms of second per iter (s/iter). The lower, the better.
训练速度如下表，指标为每次迭代的时间，以秒为单位，越低越快。

| Implementation | PSPNet (s/iter) | DeepLabV3+ (s/iter) |
|----------------|-----------------|---------------------|
Expand All @@ -160,4 +149,4 @@ The training speed is reported as followed, in terms of second per iter (s/iter)
| [CASILVision](https://github.com/CSAILVision/semantic-segmentation-pytorch) | 1.15 | N/A |
| [vedaseg](https://github.com/Media-Smart/vedaseg) | 0.95 | 1.25 |

Note: The output stride of DeepLabV3+ is 8.
注意：DeepLabV3+ 的输出步长为 8。

0 comments on commit 1105422

Please sign in to comment.