when i use this to generate scene graph,i get "RuntimeError", and but some it's fine working on other graph
To Reproduce
Steps to reproduce the behavior:
1.i use this command from "readme"
python tools/demo/demo_image.py --config_file sgg_configs/vrd/R152FPN_vrd_reldn.yaml \
--img_file /home/user/JL/myhome/datasets/VQA_datasets/okvqa/Images/train2014/COCO_train2014_000000179526.jpg \
--save_file 'output/1024px-Gen_Robert_E_Lee_on_Traveler_at_Gettysburg_Pa.reldn_relation.jpg'\
--visualize_relation MODEL.ROI_RELATION_HEAD.DETECTOR_PRE_CALCULATED False
2.i just modified the "img_file" for coco 2014 train datasets like this picture id "COCO_train2014_000000179526.jpg"
and i get the error:
./datasets
openimages_v5c/vrd/ji_vrd_labelmap.json
Traceback (most recent call last):
File "tools/demo/demo_image.py", line 192, in
main()
File "tools/demo/demo_image.py", line 128, in main
dets = detect_objects_on_single_image(model, transforms, cv2_img)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/tools/demo/detect_utils.py", line 25, in detect_objects_on_single_image
prediction = model(img_input)
File "/home/user/JL/myhome/anaconda3/envs/scene_graph_benchmark/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, kwargs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/scene_parser.py", line 319, in forward
x_pairs, prediction_pairs, relation_losses = self.relation_head(features, predictions, targets)
File "/home/user/JL/myhome/anaconda3/envs/scene_graph_benchmark/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, *kwargs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/relation_head.py", line 211, in forward
= self.rel_predictor(features, proposals, proposal_pairs)
File "/home/user/JL/myhome/anaconda3/envs/scene_graph_benchmark/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(input, kwargs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/reldn/reldn.py", line 80, in forward
xpred, = self.pred_feature_extractor(features, proposals, proposal_pairs, use_relu=False)
File "/home/user/JL/myhome/anaconda3/envs/scene_graph_benchmark/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/roi_relation_feature_extractors.py", line 134, in forward
rel_inds = _get_rel_inds(im_inds, im_inds_pairs, proposal_idx_pairs, len(proposals))
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/sparse_targets.py", line 113, in _get_rel_inds
num_rels_im = full_count(im_inds_pairs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/sparse_targets.py", line 96, in full_count
num_im_new[unique_idx] = num_im
RuntimeError: Index put requires the source and destination dtypes match, got Float for the destination and Long for the source.
Expected behavior
i want to use this to generate scence graph for coco datasets
wget https://raw.githubusercontent.com/pytorch/pytorch/master/torch/utils/collect_env.py
# For security purposes, please check the contents of collect_env.py before running it.
python collect_env.py
PyTorch Version (e.g., 1.0):1.7
OS (e.g., Linux):centos 7
How you installed PyTorch (conda, pip, source):conda
Build command you used (if compiling from source):python setup.py install
Python version:3.8.12
CUDA/cuDNN version:10.2
GPU models and configuration: 4x NVIDIA GeForce 2080ti
🐛 Bug
when i use this to generate scene graph,i get "RuntimeError", and but some it's fine working on other graph
To Reproduce
Steps to reproduce the behavior:
1.i use this command from "readme" python tools/demo/demo_image.py --config_file sgg_configs/vrd/R152FPN_vrd_reldn.yaml \ --img_file /home/user/JL/myhome/datasets/VQA_datasets/okvqa/Images/train2014/COCO_train2014_000000179526.jpg \ --save_file 'output/1024px-Gen_Robert_E_Lee_on_Traveler_at_Gettysburg_Pa.reldn_relation.jpg'\ --visualize_relation MODEL.ROI_RELATION_HEAD.DETECTOR_PRE_CALCULATED False 2.i just modified the "img_file" for coco 2014 train datasets like this picture id "COCO_train2014_000000179526.jpg"
./datasets openimages_v5c/vrd/ji_vrd_labelmap.json Traceback (most recent call last): File "tools/demo/demo_image.py", line 192, in
main()
File "tools/demo/demo_image.py", line 128, in main
dets = detect_objects_on_single_image(model, transforms, cv2_img)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/tools/demo/detect_utils.py", line 25, in detect_objects_on_single_image
prediction = model(img_input)
File "/home/user/JL/myhome/anaconda3/envs/scene_graph_benchmark/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, kwargs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/scene_parser.py", line 319, in forward
x_pairs, prediction_pairs, relation_losses = self.relation_head(features, predictions, targets)
File "/home/user/JL/myhome/anaconda3/envs/scene_graph_benchmark/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, *kwargs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/relation_head.py", line 211, in forward
= self.rel_predictor(features, proposals, proposal_pairs)
File "/home/user/JL/myhome/anaconda3/envs/scene_graph_benchmark/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(input, kwargs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/reldn/reldn.py", line 80, in forward
xpred, = self.pred_feature_extractor(features, proposals, proposal_pairs, use_relu=False)
File "/home/user/JL/myhome/anaconda3/envs/scene_graph_benchmark/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/roi_relation_feature_extractors.py", line 134, in forward
rel_inds = _get_rel_inds(im_inds, im_inds_pairs, proposal_idx_pairs, len(proposals))
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/sparse_targets.py", line 113, in _get_rel_inds
num_rels_im = full_count(im_inds_pairs)
File "/home/user/JL/myhome/juyterNotebook_folder/test/test_for_code/scene_graph_benchmark/scene_graph_benchmark/relation_head/sparse_targets.py", line 96, in full_count
num_im_new[unique_idx] = num_im
RuntimeError: Index put requires the source and destination dtypes match, got Float for the destination and Long for the source.
Expected behavior
i want to use this to generate scence graph for coco datasets
Environment
Please copy and paste the output from the environment collection script from PyTorch (or fill out the checklist below manually).
You can get the script and run it with:
conda,pip, source):condaAdditional context