Glorot normal initializer, also called Xavier normal initializer. It draws samples from a normal distribution centered on 0 with `stddev = sqrt(2 / (fan_in + fan_out))` where `fan_in` is the number of input units in the weight tensor and `fan_out` is the number of output units in th
(t)
| 95 | |
| 96 | |
| 97 | def glorot_normal(t): |
| 98 | """Glorot normal initializer, also called Xavier normal initializer. |
| 99 | |
| 100 | It draws samples from a normal distribution centered on 0 |
| 101 | with `stddev = sqrt(2 / (fan_in + fan_out))` |
| 102 | where `fan_in` is the number of input units in the weight tensor |
| 103 | and `fan_out` is the number of output units in the weight tensor. |
| 104 | |
| 105 | # Arguments |
| 106 | t(Tensor):the tensor to be filled in. |
| 107 | |
| 108 | # References |
| 109 | - [Understanding the difficulty of training deep feedforward neural |
| 110 | networks](http://jmlr.org/proceedings/papers/v9/glorot10a/glorot10a.pdf) |
| 111 | """ |
| 112 | _random_fill(t, scale=1., mode='fan_avg', distribution='normal') |
| 113 | |
| 114 | |
| 115 | def glorot_uniform(t): |
nothing calls this directly
no test coverage detected