Struct EmbeddingFromPretrainedOptions

• Defined in File embedding.h

Struct Documentation

struct EmbeddingFromPretrainedOptions

Options for the Embedding::from_pretrained function.

Public Functions

inline auto freeze(const bool &new_freeze) -> decltype(*this)

If true, the tensor does not get updated in the learning process.

Equivalent to embedding.weight.requires_grad_(false). Default: true

inline auto freeze(bool &&new_freeze) -> decltype(*this)

inline const bool &freeze() const noexcept

inline bool &freeze() noexcept

inline auto padding_idx(const std::optional<int64_t> &new_padding_idx) -> decltype(*this)

If specified, the entries at padding_idx do not contribute to the gradient; therefore, the embedding vector at padding_idx is not updated during training, i.e.

it remains as a fixed “pad”.

inline auto padding_idx(std::optional<int64_t> &&new_padding_idx) -> decltype(*this)

inline const std::optional<int64_t> &padding_idx() const noexcept

inline std::optional<int64_t> &padding_idx() noexcept

inline auto max_norm(const std::optional<double> &new_max_norm) -> decltype(*this)

If given, each embedding vector with norm larger than max_norm is renormalized to have norm max_norm.

inline auto max_norm(std::optional<double> &&new_max_norm) -> decltype(*this)

inline const std::optional<double> &max_norm() const noexcept

inline std::optional<double> &max_norm() noexcept

inline auto norm_type(const double &new_norm_type) -> decltype(*this)

The p of the p-norm to compute for the max_norm option. Default 2.

inline auto norm_type(double &&new_norm_type) -> decltype(*this)

inline const double &norm_type() const noexcept

inline double &norm_type() noexcept

inline auto scale_grad_by_freq(const bool &new_scale_grad_by_freq) -> decltype(*this)

If given, this will scale gradients by the inverse of frequency of the words in the mini-batch.

Default false.

inline auto scale_grad_by_freq(bool &&new_scale_grad_by_freq) -> decltype(*this)

inline const bool &scale_grad_by_freq() const noexcept

inline bool &scale_grad_by_freq() noexcept

inline auto sparse(const bool &new_sparse) -> decltype(*this)

If true, gradient w.r.t. weight matrix will be a sparse tensor.

inline auto sparse(bool &&new_sparse) -> decltype(*this)

inline const bool &sparse() const noexcept

inline bool &sparse() noexcept