I am getting tons of this warnings:
[W TensorAdvancedIndexing.cpp:1391] Warning: masked_fill_ received a mask with dtype torch.uint8, this behavior is now deprecated,please use a mask with dtype torch.bool instead. (function masked_fill_impl_cpu)
my mask is specified as tensor:kBool type, but it keeps complaining about it.
how can I fix it ? please help.
my code as below:
auto inputs = at::from_blob(feats.data(), {1, num}, at::kFloat).reshape({1, frame_count, dimension_size});
torch::Tensor length = torch::tensor({frame_count}, {torch::kInt64});
auto mask = torch::ones({1, frame_count}, torch::kBool);
//auto options = torch::TensorOptions().dtype(torch::kBool);
// auto mask = torch::ones({1, frame_count}, options);
std::vector<torch::jit::IValue> feat_inputs;
feat_inputs.emplace_back(inputs);
feat_inputs.emplace_back(length);
feat_inputs.emplace_back(mask);
torch::Tensor outputs = module.forward(feat_inputs).toTensor();
or how can i disable this warning? it complicates the log with useless messages.