Academia.eduAcademia.edu

Figure 2 – uploaded by Shreyansh singh

See full PDFdownloadDownload figure
Table 2. Results on the Adult dataset Table 3. Results on the News dataset

Table 2 Results on the Adult dataset Table 3. Results on the News dataset

Related Figures (4)

Fig. 1. Architecture of the generator (left) and discriminator (right) of MeTGAN. indicates concatenation. The input z to the generator is the noise vector sample from a standard multivariate normal distribution. cond is the conditional vector tha represents the column name and value that the generator conditions upon. For the di: criminator, a PacGAN [8] framework with n samples in each pac is used. r1,...,7n is th real or generated data which is passed through the discriminator, and cond), ..., cond, is used for the conditional vector. The outputs (, a indicate the one-hot vector of th mode and the value within the mode for continuous columns and d is the one-hot vecto of the discrete values for categorical columns. The Score from the discriminator is th critic score indicating the realness of the input data given the conditional vector.
Fig. 1. Architecture of the generator (left) and discriminator (right) of MeTGAN. indicates concatenation. The input z to the generator is the noise vector sample from a standard multivariate normal distribution. cond is the conditional vector tha represents the column name and value that the generator conditions upon. For the di: criminator, a PacGAN [8] framework with n samples in each pac is used. r1,...,7n is th real or generated data which is passed through the discriminator, and cond), ..., cond, is used for the conditional vector. The outputs (, a indicate the one-hot vector of th mode and the value within the mode for continuous columns and d is the one-hot vecto of the discrete values for categorical columns. The Score from the discriminator is th critic score indicating the realness of the input data given the conditional vector.
Table 4. Results on the Loan dataset
Table 4. Results on the Loan dataset
Table 5. GPU Memory usage on the Loan dataset
Table 5. GPU Memory usage on the Loan dataset
Related topics:
Computer ScienceCategorical Latent Variable Modeling
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved