无法运行test.py，使用官方例子图片，提示(需要整数元组)错误TypeError: conv_transpose2d(): argument 'dilation' (position 8) must be tuple of ints, not tuple

[evanlin88]

@lcy0604

1. 下载数据集示例(只留1.jpg文件)到test_img，并执行 python flist.py --path test_img/ --output ./datasets/enstext_test_img.flis 成功执行： loading test_img/ loading test_img/structure_im loading test_img/all_images loading test_img/all_gts loading test_img/all_labels

2. 执行CUDA_VISIBLE_DEVICES=0 python test.py --bs 1 --gpus 1 --img_flist datasets/enstext_test_img.flist --model models/CTRNet_model_final.pt --save_path ./results --save 执行效果：

/data/ctrnet/CTRNet-master/dataset.py:270: RuntimeWarning: invalid value encountered in divide cosin = (square_distance - square_distance_1 - square_distance_2) / \

flist: datasets/enstext_test_img.flist str flist: datasets/enstext_test_img.flist isfile: ['test_img/all_images/1.jpg' 'test_img/all_labels/1.jpg' 'test_img/structure_im/1.jpg'] Total instance number: 3 ===> Loaded datasets ....

TypeError: conv_transpose2d(): argument 'dilation' (position 8) must be tuple of ints, not tuple

是否可以提供一下依赖包版本？requirement.txt文件，下面是当前环境的版本 Package Version

---

anyio 4.2.0 certifi 2023.11.17 charset-normalizer 3.3.2 exceptiongroup 1.2.0 h11 0.14.0 httpcore 1.0.2 httpx 0.26.0 idna 3.6 imageio 2.33.1 inplace-abn 1.1.0 lazy_loader 0.3 networkx 3.1 numpy 1.24.4 opencv-python 4.9.0.80 packaging 23.2 pillow 10.2.0 pip 23.3.1 polygon 1.1.5 Polygon3 3.0.9.1 pyclipper 1.3.0.post5 PyWavelets 1.4.1 requests 2.31.0 scikit-image 0.21.0 scipy 1.10.1 setuptools 68.2.2 shapely 2.0.2 sniffio 1.3.0 tifffile 2023.7.10 torch 1.9.0+cu111 torchaudio 0.9.0 torchlight 1.0 torchvision 0.10.0+cu111 typing_extensions 4.9.0 urllib3 2.1.0 websocket-client 1.7.0 websockets 12.0 wheel 0.41.2

[evanlin88]

原来是版本冲突问题，降了版本后能运行