:param path_to_data: :param actions: :param input_n: :param output_n: :param dct_used: :param split:

:param path_to_data: :param actions: :param input_n: :param output_n: :param dct_used: :param split:

Compute the 2D matrix of distances between all the embeddings. Args: embeddings: tensor of shape (batch_size, embed_dim) squared:

Compute the 2D matrix of distances between all the embeddings. Args: embeddings: tensor of shape (batch_size, embed_dim) squared:

We define some variables that are useful to run the kinematic tree Args None Returns parent: 32-long vector with parent-chil

x: [bs,..., jn, 3] or [bs,..., jn-1, 3] if invert x0: [1,..., jn, 3] parents: [-1,0,1 ...]

x: [bs,..., jn, 3] or [bs,..., jn-1, 3] if invert x0: [1,..., jn, 3] parents: [-1,0,1 ...]

Define the list of actions we are using. Args action: String with the passed action. Could be "all" Returns actions: List of

Converts expmap to quaternion batch pytorch version ported from the corresponding numpy method above :param R: N*3 :return: N*4

Converts an exponential map angle to a rotation matrix Matlab port to python for evaluation purposes I believe this is also called Rodrig

Converts expmap matrix to rotation batch pytorch version ported from the corresponding method above :param r: N*3 :return: N*3*3

convert expmaps to joint locations :param expmap: N*99 :return: N*32*3

convert expmaps to joint locations :param expmap: N*99 :return: N*32*3

Adapted from https://github.com/una-dinosauria/human-motion-prediction/blob/master/src/seq2seq_model.py#L478 which originaly from In ord

Adapted from https://github.com/una-dinosauria/human-motion-prediction/blob/master/src/seq2seq_model.py#L478 which originaly from In ord

x,x_cond: [bs,data_dim]

@param X: @return:

p3d: [bs,16,3] or [bs,48]

p3d: [bs,16,3] or [bs,48]

p3d: [bs,14,3] or [bs,42]

p3d: [bs,14,3] or [bs,42]

Also borrowed for SRNN code. Computes mean, stdev and dimensions to ignore. https://github.com/asheshjain399/RNNexp/blob/srnn/structural_rnn/

Normalize input data by removing unused dimensions, subtracting the mean and dividing by the standard deviation Args data: nx99 ma

Converts the output of the neural network to a format that is more easy to manipulate for, e.g. conversion to other format or visualization

Converts a rotation matrix to Euler angles Matlab port to python for evaluation purposes https://github.com/asheshjain399/RNNexp/blob/srn

Converts a rotation matrix to euler angles batch pytorch version ported from the corresponding numpy method above :param R:N*3*3 :re

Converts a rotation matrix to quaternion batch pytorch version ported from the corresponding numpy method above :param R: N * 3 * 3 :

@param gt_pos: seq_l x jn x 3 @param pre_pos: n seq_l x jn x 3 @param labels: n @param comments: @return:

@param gt_pos: seq_l x jn x 3 @param pre_pos: n seq_l x jn x 3 @param labels: n @param comments: @return: