Args: enc_inputs: [batch_size, src_len] dec_inputs: [batch_size, tgt_len]

Args: enc_inputs: [batch_size, src_len] dec_inputs: [batch_size, tgt_len]

Controls whether events will be forwarded to listeners_. Set to false in death test child processes.

r""" freeze the weights of the base model Args: base_model: the base model

Decompress floating-point number. See also:** Documentation for [std::frexp](https://en.cppreference.com/w/cpp/numeric/math/frexp). \param arg number

Init a General Matrix multiplication(Gemm) operator. Compute `Y = alpha * A' * B' + beta * C`, where input tensor A has shape (M, K) or (K, M

Only for the MLP Returns:

Generate singa_config.h file to define some macros for the cpp code. Args: with_cuda(bool): indicator for cudnn and cuda lib with

cudnn requires stride dimensions to conform to the format of the shape input as well 1.) Stride dimensions to be set to a minimum of 4 for 4d an

Project the tokenized prediction back to the original text.

return output shape of conv2d or pooling, ! borrow from onnx Args: auto_pad: string input_spatial_shape: list[int]

return padding shape of conv2d or pooling, Args: pad_mode: string kernel_spatial_shape: list[int] strides_spatial: li

Get parameters of this layer and all sublayers Returns: parameters(dict): A dictionary contains parameter names and