When reproducing on collab, facing issue with the scene_graph file

[kxusx]

ERROR FACED :
No. of images available: 43728. Please download the dataset following the instructions.
Traceback (most recent call last):
File "/content/fcsgg/tools/train_net.py", line 43, in
from fcsgg.config import add_fcsgg_config
File "/content/fcsgg/fcsgg/init.py", line 1, in
from .modeling import meta_arch
File "/content/fcsgg/fcsgg/modeling/init.py", line 1, in
from .meta_arch import CenterNet
File "/content/fcsgg/fcsgg/modeling/meta_arch/init.py", line 1, in
from .onestage_detector import CenterNet
File "/content/fcsgg/fcsgg/modeling/meta_arch/onestage_detector.py", line 35, in
from fcsgg.data.detection_utils import GroundTruthGen
File "/content/fcsgg/fcsgg/data/init.py", line 2, in
from . import datasets
File "/content/fcsgg/fcsgg/data/datasets/init.py", line 2, in
from .visual_genome import register_visual_genome
File "/content/fcsgg/fcsgg/data/datasets/visual_genome.py", line 500, in
register_visual_genome()
File "/content/fcsgg/fcsgg/data/datasets/visual_genome.py", line 494, in register_visual_genome
predicate_stats=vg_parser.get_predicate_stats(),
File "/content/fcsgg/fcsgg/data/datasets/visual_genome.py", line 365, in get_predicate_stats
roi_h5 = h5py.File(self.roidb_file, 'r')
File "/usr/local/lib/python3.10/dist-packages/h5py/_hl/files.py", line 567, in init
fid = make_fid(name, mode, userblock_size, fapl, fcpl, swmr=swmr)
File "/usr/local/lib/python3.10/dist-packages/h5py/_hl/files.py", line 231, in make_fid
fid = h5f.open(name, flags, fapl=fapl)
File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper
File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper
File "h5py/h5f.pyx", line 106, in h5py.h5f.open
OSError: Unable to open file (truncated file: eof = 9437184, sblock->base_addr = 0, stored_eof = 150724042)

I re-downloaded the h5 file and tried again still same issue. might be corrupted

[liuhengyue]

I saw your assertion failed as "No. of images available: 43728. Please download the dataset following the instructions." It seems that you have some images but not all. The file is not liked to be corrupted as it works on my end.

1. check if you have the correct path and the whole dataset; 2. try an old environment 'cause I see you are using Python 3.10, maybe use older versions of python, pytorch, h5py, etc.