关于训练自己数据集的问题

PaddlePaddle / PaddleSeg

Easy-to-use image segmentation library with awesome pre-trained model zoo, supporting wide-range of practical tasks in Semantic Segmentation, Interactive Segmentation, Panoptic Segmentation, Image Matting, 3D Segmentation, etc.

https://arxiv.org/abs/2101.06175

Apache License 2.0

8.73k stars 1.68k forks source link

关于训练自己数据集的问题 #963

Closed VJoer closed 1 year ago

VJoer commented 3 years ago

我按照教程训练制作自己的数据集，使用Fast scnn 进行多分类语义分割训练，配置文件也是按照教程来的，但是训练结果只是二分类，请问是什么原因。 train_dataset: type: Dataset dataset_root: data train_path: data/train.txt num_classes: 25 transforms:

type: ResizeStepScaling min_scale_factor: 0.5 max_scale_factor: 2.0 scale_step_size: 0.25
type: RandomPaddingCrop crop_size: [512, 512]
type: RandomHorizontalFlip
type: Normalize mode: train

michaelowenliu commented 3 years ago

@VJoer Hi, make sure the num_classes in the model is correct. https://github.com/PaddlePaddle/PaddleSeg/blob/release/v2.0/paddleseg/models/fast_scnn.py#L61

VJoer commented 3 years ago

@VJoer Hi, make sure the num_classes in the model is correct. https://github.com/PaddlePaddle/PaddleSeg/blob/release/v2.0/paddleseg/models/fast_scnn.py#L61

好的，请问配置文件中如何修改传入图片的大小

michaelowenliu commented 3 years ago

@VJoer Hi, please take a look at our config files.

MansGod commented 3 years ago

您好，我也是用fastscnn训练，请问您预训练模型的参数怎么配置的呢，我修改了pretrained_model的路径，但是训练时报错无法在预训练模型中找到权重

michaelowenliu commented 3 years ago

@MansGod Hi, could you please put the error log?

github-actions[bot] commented 1 year ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 7 days if no further activity occurs. Thank you for your contributions.