* @brief Construct a memory space * @param _device_id GPU id, -1 for CPU * @param _count number of data * @param _stream CUDA stream

* Construct a general edge sampler * @param _solver pointer to graph embedding solver * @param _device_id GPU id */

* @brief Construct a general solver * @param device_ids list of GPU ids, {} for auto * @param _num_sampler_per_worker number of sampler thre

Default constructor */

* Construct a general training worker * @param _solver pointer to graph embedding solver * @param _device_id GPU id */

Rotate learn 3D coordinates as an animation. Parameters: Y (list of str, optional): labels of vectors file_n

Determine and allocate all resources for the sampler */

Determine and allocate all resources for the worker */

Should build alias reference for embeddings */

Build the alias table for negative sampling */

Monitor the arguments of a function. Parameters: function (function): function to monitor Returns:

Free GPU memory */

Free GPU memory */

Constant schedule */

Count edges for each sample block. This function can be parallelized. */

Distribute edges to the sample pool. This function can be parallelized. */

Distribute edges to the sample pool. This function can be parallelized. */

Predict the distribution of missing entity or relation for triplets. Parameters: H (list of str, optional): names of hea

Fill the memory space with data. Automatically resize the memory when necessary. */

Determine the minimum number of partitions */

Determine and prepare the sampling function */

Determine the schedule of partitions */

* @param protocols protocols of embeddings * @param shapes shapes of embeddings * @param sampler_protocol protocol of negative sampling

@return GPU memory cost */