Sentence examples for compute the dataset from inspiring English sources

HotML is designed in the parameter server (PS) architecture where the servers manage the globally shared parameters organized in tabular structure, and the workers compute the dataset in parallel and update the global parameters.

We observed that the total time taken to compute the whole dataset when increasing the number of processors available for computing appears to follow a power law distribution.

In the second experiment we similarly measured the time to compute the TRYP dataset while increasing the number of processors (Fig.  3b).

POTION was able to compute the TRYP dataset in approximately 129 min with a single processor (7744 s) and took approximately 26 and 15 min (1581 and 942 s, respectively) to compute the same dataset with 5 and 10 processors, respectively, indicating that POTION processed the entire dataset 4.9 and 8.2 times faster with 5 and 10 processors, respectively.

We also observed that the total time taken to compute the whole dataset to follow a power law distribution when increasing the number of processors (slope of linear regression for log-log plot: −0.95) in a pattern similar to the one observed in MYC dataset (Fig.  3a).

POTION took approximately 1 day (~25 h, 89,563 s) to compute the entire MYC dataset using 90 processors, and the linear time to compute this entire dataset with a single processor was calculated to be more than 2 months (~1547 h, 5,568,246 s), indicating that POTION processed an entire dataset of genome-scale data 62 times faster when in multi-processor mode.

We then used this technique to compute the five imputed datasets.

By using trivially parallel computing the entire dataset could be processed in a single weekend.

For each model, the parameters (including the topology and the branch lengths) were learnt on one of the two datasets, and used to compute the likelihood of the other dataset.

We trained the fusion model with the scores computed over the training dataset, and we used the test dataset to compute the SER.

Most existing covariate shift methods compute the resampling weight of training dataset and then train a weighted risk model to predict on test dataset.