Given random variable X, the cumulative distribution function cdf is: tfd_log_cdf(x) := Log[ P[X <= x] ] Often, a numerical approximation can be used for tfd_log_cdf(x) that yields a more accurate answer than simply taking the logarithm of the cdf when x << -1.

tfd_log_cdf(distribution, value, ...)

Arguments

distribution

The distribution being used.

value

float or double Tensor.

...

Additional parameters passed to Python.

Value

a Tensor of shape sample_shape(x) + self$batch_shape with values of type self$dtype.

See also

Examples

# \donttest{ d <- tfd_normal(loc = c(1, 2), scale = c(1, 0.5)) x <- d %>% tfd_sample() d %>% tfd_log_cdf(x)
#> tf.Tensor([-0.05474529 -1.1571263 ], shape=(2,), dtype=float32)
# }