Skip to content

[FasterTransformer] Fix bug of trt sample. #501

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
byshiue merged 1 commit into from
May 13, 2020
Merged

[FasterTransformer] Fix bug of trt sample. #501

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions FasterTransformer/v2/fastertransformer/trt_plugin/trt_model.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,8 +44,8 @@ class TRT_Transformer
check_cuda_error(cudaFree(buffers[input_index_]));
check_cuda_error(cudaFree(buffers[mask_index_]));
check_cuda_error(cudaFree(buffers[output_index_]));
engine_->destroy();
context_->destroy();
engine_->destroy();
}

nvinfer1::Weights point2weight(T* ptr, int size)
Expand All @@ -65,7 +65,7 @@ class TRT_Transformer
auto from_tensor = network->addInput(INPUT_BLOB_NAME, dtype_, nvinfer1::Dims2{seq_len_, hidden_dim_});
auto mask_tensor = network->addInput(MASK_BLOB_NAME, dtype_, nvinfer1::Dims2{seq_len_, seq_len_});

assert(input_tensor);
assert(from_tensor);
assert(mask_tensor);

nvinfer1::ITensor* output_tensor = nullptr;
Expand Down
1 change: 0 additions & 1 deletion FasterTransformer/v2/sample/tensorRT/transformer_trt.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -130,7 +130,6 @@ void run_bert_transformer(int batch_size, int seq_len, int layers, int head_num,

TRT_Transformer<T>* trt_transformer = new TRT_Transformer<T>(batch_size, seq_len, head_num, hidden_dim, layers);
trt_transformer->build_engine(params);

trt_transformer->do_inference(batch_size, h_from_tensor, h_attr_mask, h_transformer_out, stream);

delete trt_transformer;
Expand Down