I am calculating word similarity using torch::Embedding module by pretrained wordvector (glove.300d) on Ubuntu 18.04LTS PyTorch C++ (1.5.1, CUDA 10.1). I believe I have moved everything I can to the GPU, but when I execute it, it still says (full error log on the end of the question):
Expected object of device type cuda but got device type cpu for
argument #1 'self' in call to _th_index_select
(checked_dense_tensor_unwrap at /pytorch/aten/src/ATen/Utils.h:72)
I have checked my model initialization method in main.cpp, and it is okay if I only do initialization.
SimilarityModel simiModel(args, 400000, 300);
simiModel.to(device);
//model forward
torch::Tensor data = ids.index({Slice(i*batch_size, (i+1)*batch_size), Slice()}).to(torch::kInt64).to(device); //take a batch
tie(score, indice) = simiModel.forward(data); //forward and transfer score, indice to cpu for further calculation
and this is how I define SimilarityModel in Similarity.h:
class SimilarityModel : public torch::nn::Module {
public:
int64_t topk; // num of top words;
Dictionary dict;
int64_t vocab_size;
int64_t embedding_dim;
torch::nn::Embedding embedding{nullptr};
vector<vector<float> > vec_embed;
SimilarityModel(unordered_map<string, string> args, int64_t vocab_size, int64_t embed_dim);
tuple<torch::Tensor, torch::Tensor> forward(torch::Tensor x);
};
At the same time I have done embedding initialization in SimilarityModel function in Similarity.cpp:
SimilarityModel::SimilarityModel(unordered_map<string, string> args, int64_t vocab_size, int64_t embed_dim)
:embedding(vocab_size, embed_dim) { //Embedding initialize
this->topk = stoi(args["topk"]);
vector<vector<float> > pre_embed;
tie(pre_embed, dict) = loadwordvec(args); //load pretrained wordvec from txt file
this->vocab_size = int64_t(dict.size());
this->embedding_dim = int64_t(pre_embed[0].size());
this->vec_embed = pre_embed;
this->dict = dict;
vector<float> temp_embed;
for(const auto& i : pre_embed) //faltten to 1-d
for(const auto& j : i)
temp_embed.push_back(j);
torch::Tensor data = torch::from_blob(temp_embed.data(), {this->vocab_size, this->embedding_dim}, torch::TensorOptions().dtype(torch::kFloat32)).clone(); //vector to tensor
register_module("embedding", embedding);
this->embedding = embedding.from_pretrained(data, torch::nn::EmbeddingFromPretrainedOptions().freeze(true));
}
and forward function in Similarity.cpp:
tuple<torch::Tensor, torch::Tensor> SimilarityModel::forward(torch::Tensor x) {
auto cuda_available = torch::cuda::is_available(); //copy to gpu
torch::Device device(cuda_available ? torch::kCUDA : torch::kCPU);
torch::Tensor wordvec;
wordvec = this->embedding->forward(x).to(device); //python:embedding(x)
torch::Tensor similarity_score = wordvec.matmul(this->embedding->weight.transpose(0, 1)).to(device);
torch::Tensor score, indice;
tie(score, indice) = similarity_score.topk(this->topk, -1, true, true); //Tensor.topk(int64_t k, int64_t dim, bool largest = true, bool sorted = true)
score = score.to(device);
indice = indice.to(device);
score.slice(1, 1, score.size(1)); //Tensor.slice(int64_t dim, int64_t start, int64_t end, int64_t step)
indice.slice(1, 1, indice.size(1));
return {score.cpu(), indice.cpu()}; //transfer to cpu for further calculation
}
As for intermediate variables in forward() have been put to GPU too. However, I totally have no idea which one is left in CPU, and the error log does not help so much. I have tried the method in Expected object of device type cuda but got device type cpu for argument #1 'self' in call to _th_index_select to do SimilarityModel().to(device), but that does not work. I am still having a hard time reading this error log and would like some instructions on how to debug such questions.
Error log:
terminate called after throwing an instance of 'c10::Error'
what(): Expected object of device type cuda but got device type cpu for argument #1 'self' in call to _th_index_select (checked_dense_tensor_unwrap at /pytorch/aten/src/ATen/Utils.h:72)
frame #0: c10::Error::Error(c10::SourceLocation, std::string const&) + 0x46 (0x7fb566a27536 in /home/switchsyj/Downloads/libtorch/lib/libc10.so)
frame #1: <unknown function> + 0x101a80b (0x7fb520fa380b in /home/switchsyj/Downloads/libtorch/lib/libtorch_cuda.so)
frame #2: <unknown function> + 0x105009c (0x7fb520fd909c in /home/switchsyj/Downloads/libtorch/lib/libtorch_cuda.so)
frame #3: <unknown function> + 0xf9d76b (0x7fb520f2676b in /home/switchsyj/Downloads/libtorch/lib/libtorch_cuda.so)
frame #4: <unknown function> + 0x10c44e3 (0x7fb558d224e3 in /home/switchsyj/Downloads/libtorch/lib/libtorch_cpu.so)
frame #5: at::native::embedding(at::Tensor const&, at::Tensor const&, long, bool, bool) + 0x2e2 (0x7fb558870712 in /home/switchsyj/Downloads/libtorch/lib/libtorch_cpu.so)
frame #6: <unknown function> + 0x114ef9d (0x7fb558dacf9d in /home/switchsyj/Downloads/libtorch/lib/libtorch_cpu.so)
frame #7: <unknown function> + 0x1187b4d (0x7fb558de5b4d in /home/switchsyj/Downloads/libtorch/lib/libtorch_cpu.so)
frame #8: <unknown function> + 0x2bfe42f (0x7fb55a85c42f in /home/switchsyj/Downloads/libtorch/lib/libtorch_cpu.so)
frame #9: <unknown function> + 0x1187b4d (0x7fb558de5b4d in /home/switchsyj/Downloads/libtorch/lib/libtorch_cpu.so)
frame #10: <unknown function> + 0x32b63a9 (0x7fb55af143a9 in /home/switchsyj/Downloads/libtorch/lib/libtorch_cpu.so)
frame #11: torch::nn::EmbeddingImpl::forward(at::Tensor const&) + 0x71 (0x7fb55af127b1 in /home/switchsyj/Downloads/libtorch/lib/libtorch_cpu.so)
frame #12: SimilarityModel::forward(at::Tensor) + 0xa9 (0x55c96b8e5793 in ./demo)
frame #13: main + 0xaba (0x55c96b8bfe5c in ./demo)
frame #14: __libc_start_main + 0xe7 (0x7fb51edf5b97 in /lib/x86_64-linux-gnu/libc.so.6)
frame #15: _start + 0x2a (0x55c96b8bd74a in ./demo)
Aborted (core dumped)
Based on the error message, one of the two following Tensors are not in the GPU when you're running SimilarityModel::forward():
this->embedding->weightxGiven that the error points to the argument #1, I'd say that weight is the one on the CPU.
Here's the call for index.select:
Tensor embedding(const Tensor & weight, const Tensor & indices,
int64_t padding_idx, bool scale_grad_by_freq, bool sparse) {
auto indices_arg = TensorArg(indices, "indices", 1);
checkScalarType("embedding", indices_arg, kLong);
// TODO: use tensor.index() after improving perf
if (indices.dim() == 1) {
return weight.index_select(0, indices);
}
auto size = indices.sizes().vec();
for (auto d : weight.sizes().slice(1)) {
size.push_back(d);
}
return weight.index_select(0, indices.reshape(-1)).view(size);
}
First, try to directly move the weight to the GPU. If it works, it means that when you called TORCH_MODULE(SimilarityModel), and moved the model to the device, it should have worked too. Remember that you have to change the name to SimilarityModelImpl (Name+Impl) in this case. Otherwise, it won't work as well.