@ -1316,7 +1316,7 @@ def dropout(x,
1. downgrade_in_infer(default), downgrade the outcome at inference
- train: out = input * mask
- inference: out = input * (1 - dropout_prob)
- inference: out = input * (1.0 - dropout_prob)
(mask is a tensor same shape with input, value is 0 or 1
ratio of 0 is dropout_prob)