about s3dis dataset

[MRCHENWJ]

Hello, I encountered an issue during preprocessing S3DIS data, how can I solve it?

(isbnet) root@autodl-container-58ce11bf52-76ea3b1f:~/autodl-tmp/ISBNet-master/dataset/s3dis# python prepare_s3dis.py Traceback (most recent call last): File "prepare_s3dis.py", line 158, in preprocess_s3dis(cfg.data_dir) File "prepare_s3dis.py", line 143, in preprocess_s3dis scene_data = read_scene_txt(scene_name, data_dir) File "prepare_s3dis.py", line 120, in read_scene_txt pts = np.loadtxt(os.path.join(data_dir + "/" + area, name, name + ".txt")) File "/root/miniconda3/envs/isbnet/lib/python3.7/site-packages/numpy/lib/npyio.py", line 1148, in loadtxt for x in read_data(_loadtxt_chunksize): File "/root/miniconda3/envs/isbnet/lib/python3.7/site-packages/numpy/lib/npyio.py", line 999, in read_data items = [conv(val) for (conv, val) in zip(converters, vals)] File "/root/miniconda3/envs/isbnet/lib/python3.7/site-packages/numpy/lib/npyio.py", line 999, in items = [conv(val) for (conv, val) in zip(converters, vals)] File "/root/miniconda3/envs/isbnet/lib/python3.7/site-packages/numpy/lib/npyio.py", line 736, in floatconv return float(x) ValueError: could not convert string to float: '167.\x1000000'

[Lizhinwafu]

Manually modify data

[robertod]

Hi @MRCHENWJ,

There are a few errors in the S3DIS dataset. To get the dataset working with the current version of ISBNet you need to fix the following erroneous characters:

• Line 180389 of Stanford3dDataset_v1.2_Aligned_Version\Area_5\hallway_6\Annotations\ceiling_1.txt
• Line 741101 of Stanford3dDataset_v1.2_Aligned_Version\Area_2\auditorium_1\auditorium_1.txt
• Line 926337 of Stanford3dDataset_v1.2_Aligned_Version\Area_3\hallway_2\hallway_2.txt

You can use either the lines specified or the following regex [\x00-\x1F] to find these mistakes.

You will also have to remove the underscore in the file name Stanford3dDataset_v1.2_Aligned_Version/Area_6/copyRoom_1/copy_Room_1.txt such that it reads copyRoom_1.txt.

Best, Roberto