RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation

[ghzmwhdk777]

start training...

Training epoch: 1
/usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py:92: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:141.)
img = torch.from_numpy(np.array(pic, np.float32, copy=False))
/usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py:92: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:141.)
img = torch.from_numpy(np.array(pic, np.float32, copy=False))
/usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py:92: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:141.)
img = torch.from_numpy(np.array(pic, np.float32, copy=False))
/usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py:92: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:141.)
img = torch.from_numpy(np.array(pic, np.float32, copy=False))
Traceback (most recent call last):
File "main.py", line 120, in
main(mode=1)
File "main.py", line 50, in main
model.train()
File "/content/drive/MyDrive/Colab Notebooks/SR/edge-informed-sisr-master/src/edge_match.py", line 133, in train
self.sr_model.backward(gen_loss, dis_loss)
File "/content/drive/MyDrive/Colab Notebooks/SR/edge-informed-sisr-master/src/models.py", line 283, in backward
gen_loss.backward()
File "/usr/local/lib/python3.6/dist-packages/torch/tensor.py", line 221, in backward
torch.autograd.backward(self, gradient, retain_graph, create_graph)
File "/usr/local/lib/python3.6/dist-packages/torch/autograd/init.py", line 132, in backward
allow_unreachable=True) # allow_unreachable flag

RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [1, 512, 4, 4]] is at version 2; expected version 1 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).

---

somebody know this error and answer??

[jakubcaputa]

Hey, did you solve this issue?

[zhangzheng2242]

开始训练...

训练时期：1 /usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py:92：UserWarning：给定的 NumPy 数组不可写，PyTorch 不支持不可写张量。这意味着您可以使用张量写入底层（假定不可写）NumPy 数组。您可能希望复制数组以保护其数据或使其可写，然后再将其转换为张量。对于该程序的其余部分，将禁止此类警告。（在 /pytorch/torch/csrc/utils/tensor_numpy.cpp:141 内部触发。） img = torch.from_numpy(np.array(pic, np.float32, copy=False)) /usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py:92：用户警告：给定的 NumPy 数组不可写，并且 PyTorch 不支持不可写张量。这意味着您可以使用张量写入底层（假定不可写）NumPy 数组。您可能希望复制数组以保护其数据或使其可写，然后再将其转换为张量。对于该程序的其余部分，将禁止此类警告。（在 /pytorch/torch/csrc/utils/tensor_numpy.cpp:141 内部触发。） img = torch.from_numpy(np.array(pic, np.float32, copy=False)) /usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py:92：用户警告：给定的 NumPy 数组不可写，并且 PyTorch 不支持不可写张量。这意味着您可以使用张量写入底层（假定不可写）NumPy 数组。您可能希望复制数组以保护其数据或使其可写，然后再将其转换为张量。对于该程序的其余部分，将禁止此类警告。（在 /pytorch/torch/csrc/utils/tensor_numpy.cpp:141 内部触发。） img = torch.from_numpy(np.array(pic, np.float32, copy=False)) /usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py:92：用户警告：给定的 NumPy 数组不可写，并且 PyTorch 不支持不可写张量。这意味着您可以使用张量写入底层（假定不可写）NumPy 数组。您可能希望复制数组以保护其数据或使其可写，然后再将其转换为张量。对于该程序的其余部分，将禁止此类警告。（在 /pytorch/torch/csrc/utils/tensor_numpy.cpp:141 内部触发。） img = torch.from_numpy(np.array(pic, np.float32, copy=False)) Traceback（最近一次调用最后一次）： 文件“main.py”，第 120 行，在 main(mode=1) 文件“main.py”，第 50 行，在 main model.train() 文件“/content/drive/MyDrive/Colab Notebooks/SR/edge-informed-sisr-master/src/edge_match.py​​”，第 133 行，在火车 self.sr_model.backward(gen_loss, dis_loss) 文件“/content/drive /MyDrive/Colab Notebooks/SR/edge-informed-sisr-master/src/models.py”，第 283 行，在向后的 gen_loss.backward() 文件中“/usr/local/lib/python3.6/dist-packages/ torch/tensor.py”，第 221 行，后向 torch.autograd.backward(self, gradient, retain_graph, create_graph)文件 “/usr/local/lib/python3.6/dist-packages/torch/autograd/init .py "，第 132 行，后向 allow_unreachable=True) #allow_unreachable 标志****

RuntimeError：梯度计算所需的变量之一已被就地操作修改：[torch.cuda.FloatTensor [1, 512, 4, 4]] 为版本 2；而是预期的版本 1。提示：启用异常检测以查找未能计算其梯度的操作，使用 torch.autograd.set_detect_anomaly(True)。

有人知道这个错误并回答吗？

Hey, did you solve this issue?

Hey, did you solve this issue?