You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
my video card is "NVIDIA GeForce MX150" (laptop) with 2 Gb video RAM.
OS: ubuntu 20.04 + NVidia driver 470
I have 61 custom images with marked object on them
when I execute this simple code:
from detecto import core, utils, visualize
dataset = core.Dataset('images_to_learn/')
model = core.Model(['my_object'])
model.fit(dataset)
it fails of model.fit(dataset) with the error:
Epoch 1 of 10
Begin iterating over training dataset
0%| | 0/61 [00:00<?, ?it/s]/home/xwizard/.local/lib/python3.9/site-packages/torch/functional.py:445: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:2157.)
return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]
2%|▊ | 1/61 [00:02<02:07, 2.12s/it]
Traceback (most recent call last):
File "/home/xwizard/test/main.py", line 24, in <module>
model.fit(dataset)
File "/home/xwizard/.local/lib/python3.9/site-packages/detecto/core.py", line 505, in fit
loss_dict = self._model(images, targets)
File "/home/xwizard/.local/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/xwizard/.local/lib/python3.9/site-packages/torchvision/models/detection/generalized_rcnn.py", line 96, in forward
proposals, proposal_losses = self.rpn(images, features, targets)
File "/home/xwizard/.local/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/xwizard/.local/lib/python3.9/site-packages/torchvision/models/detection/rpn.py", line 354, in forward
proposals = self.box_coder.decode(pred_bbox_deltas.detach(), anchors)
File "/home/xwizard/.local/lib/python3.9/site-packages/torchvision/models/detection/_utils.py", line 180, in decode
pred_boxes = self.decode_single(
File "/home/xwizard/.local/lib/python3.9/site-packages/torchvision/models/detection/_utils.py", line 223, in decode_single
pred_boxes1 = pred_ctr_x - c_to_c_w
RuntimeError: CUDA out of memory. Tried to allocate 2.00 MiB (GPU 0; 1.96 GiB total capacity; 1.12 GiB already allocated; 2.88 MiB free; 1.26 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
Pytorch just takes all available memory and crashes.
The text was updated successfully, but these errors were encountered:
I'm trying basic example from
https://towardsdatascience.com/build-a-custom-trained-object-detection-model-with-5-lines-of-code-713ba7f6c0fb
my video card is "NVIDIA GeForce MX150" (laptop) with 2 Gb video RAM.
OS: ubuntu 20.04 + NVidia driver 470
I have 61 custom images with marked object on them
when I execute this simple code:
it fails of model.fit(dataset) with the error:
Pytorch just takes all available memory and crashes.
The text was updated successfully, but these errors were encountered: