Unable to run the script (run_partslip.py)

[jayaramreddy10]

Hi,

Thanks for releasing the code. I am getting below issue when I try to run partslip script with masks.

Traceback (most recent call last):
File "run_partslip.py", line 63, in
Infer(f"./data/test/{category}/{model}/pc.ply", category, model, partnete_meta[category], zero_shot=False, save_dir=f"./result_ps/{category}/{model}")
File "run_partslip.py", line 47, in Infer
masks = glip_inference(glip_demo, save_dir, part_names, sam_predictor, num_views=num_views)
File "/home/jayaram/PartSLIP2/src/glip_inference.py", line 70, in glip_inference
result, top_predictions = glip_demo.run_on_web_image(image, part_names, 0.5)
File "/home/jayaram/PartSLIP2/GLIP/maskrcnn_benchmark/engine/predictor_glip.py", line 140, in run_on_web_image
predictions = self.compute_prediction(original_image, original_caption, custom_entity)
File "/home/jayaram/PartSLIP2/GLIP/maskrcnn_benchmark/engine/predictor_glip.py", line 220, in compute_prediction
predictions = self.model(image_list, captions=[original_caption], positive_map=positive_map_label_to_token)
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/jayaram/PartSLIP2/GLIP/maskrcnn_benchmark/modeling/detector/generalized_vl_rcnn.py", line 284, in forward
proposals, proposal_losses, fused_visual_features = self.rpn(images, visual_features, targets, language_dict_features, positive_map,
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/jayaram/PartSLIP2/GLIP/maskrcnn_benchmark/modeling/rpn/vldyhead.py", line 920, in forward
proj_tokens, contrastive_logits, dot_product_logits, mlm_logits, shallow_img_emb_feats, fused_visual_features = self.head(features,
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/jayaram/PartSLIP2/GLIP/maskrcnn_benchmark/modeling/rpn/vldyhead.py", line 739, in forward
dyhead_tower = self.dyhead_tower(feat_inputs)
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/container.py", line 141, in forward
input = module(input)
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/jayaram/PartSLIP2/GLIP/maskrcnn_benchmark/modeling/rpn/vldyhead.py", line 228, in forward
next_x = [self.relu(item) for item in next_x]
File "/home/jayaram/PartSLIP2/GLIP/maskrcnn_benchmark/modeling/rpn/vldyhead.py", line 228, in
next_x = [self.relu(item) for item in next_x]
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/jayaram/PartSLIP2/GLIP/maskrcnn_benchmark/layers/dyrelu.py", line 87, in forward
y = self.fc(y).view(b, self.oup * self.exp, 1, 1)
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/container.py", line 141, in forward
input = module(input)
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/modules/linear.py", line 103, in forward
return F.linear(input, self.weight, self.bias)
File "/home/jayaram/miniconda3/envs/partslip2p38/lib/python3.8/site-packages/torch/nn/functional.py", line 1848, in linear
return torch._C._nn.linear(input, weight, bias)
RuntimeError: CUDA error: CUBLAS_STATUS_INVALID_VALUE when calling cublasSgemm( handle, opa, opb, m, n, k, &alpha, a, lda, b, ldb, &beta, c, ldc)

Below are the library versions:

OS: Ubuntu 22.04
Python: 3.8.19 (I had to go for 3.8 because I was not able to install pytorch3d with 3.9, had an issue with below import command
from pytorch3d.io import IO)
Cuda: 11.3
Torch: 1.10.0

Please let me know if you need any other details?