<summary> <para>Initialize the parameters with a gaussian distribution</para> </summary> <param name='mean'>Mean of the distribution (default: 0)</par

a a 0 1 a ||| O:1 O:1 N:2 O:1

* * \brief Default constructor */

* \brief Samples a string of words/characters from the model * \details This can be used to debug and/or have fun. Try it on * new datasets!

<summary> <para>Initializes according to 'Saxe et al. (2014) (https://arxiv.org/abs/1312.6120)'</para> <para>Initializes as a random orthonormal matri

<summary> <para>VanillaLSTM allows to create an "standard" LSTM, ie with decoupled input and forget gate and no peepholes connections</para> <para>Dur

return Expression of total loss

return Expression of total loss

* \brief Create an empty tensor */

* \brief General constructor for a Trainer * * \param m ParameterCollection to be trained * \param learning_rate Initial learning rate */

given the first character of a UTF8 block, find out how wide it is see http://en.wikipedia.org/wiki/UTF-8 for more info

Adds auxiliary output.

Adds lookup parameters with a name on an explicit device. # Arguments n - Number of lookup indices. d - Dimension of each embedding. init - Initiali

Adds parameters with a name on an explicit device. # Panics Panics if `name` has an invalid character including `/` and `_`.

Adds a subcollection with a name. # Panics Panics if `name` has an invalid character including `/` and `_`.

Retrieves internal values in the tensor as a vector. For higher order tensors this returns the flattened value.

* \brief which inputs can be batched * \detail This will be true for inputs that should be concatenated when * autobatching, and false f

* \brief create a pseudonode for autobatching * \detail This will combine together multiple nodes into one big node for * the automatic

* \brief reshape the tensors for auto * \detail Takes in info, and reshapes the dimensions of xs (for which * "concat" is true), and fx.

Returns the batch dimension. # Examples ``` # use dynet::Dim; let dim = Dim::from(([4, 5, 6], 8)); assert_eq!(dim.batch_elems(), 8); ```

Returns the size of the second dimension. This returns `1` if the dim has only one dimension or no dimensions. # Examples ``` # use dynet::Dim; let

Copies the parameters of another builder.

---

---

Dump the ComputationGraph.

* \brief Enable dropout * \details This is supposed to be used during training or during testing if you want to sample outputs using montecarlo