Instantiates a variable with values drawn from a uniform distribution. Arguments: shape: Tuple of integers, shape of returned Keras variable. low: Float, lower boundary of the output interval. high: Float, upper boundary of the output interval. dtype: String, dtype of retu
(shape, low, high, dtype=None, name=None, seed=None)
| 1437 | |
| 1438 | @keras_export('keras.backend.random_uniform_variable') |
| 1439 | def random_uniform_variable(shape, low, high, dtype=None, name=None, seed=None): |
| 1440 | """Instantiates a variable with values drawn from a uniform distribution. |
| 1441 | |
| 1442 | Arguments: |
| 1443 | shape: Tuple of integers, shape of returned Keras variable. |
| 1444 | low: Float, lower boundary of the output interval. |
| 1445 | high: Float, upper boundary of the output interval. |
| 1446 | dtype: String, dtype of returned Keras variable. |
| 1447 | name: String, name of returned Keras variable. |
| 1448 | seed: Integer, random seed. |
| 1449 | |
| 1450 | Returns: |
| 1451 | A Keras variable, filled with drawn samples. |
| 1452 | |
| 1453 | Example: |
| 1454 | ```python |
| 1455 | # TensorFlow example |
| 1456 | >>> kvar = K.random_uniform_variable((2,3), 0, 1) |
| 1457 | >>> kvar |
| 1458 | <tensorflow.python.ops.variables.Variable object at 0x10ab40b10> |
| 1459 | >>> K.eval(kvar) |
| 1460 | array([[ 0.10940075, 0.10047495, 0.476143 ], |
| 1461 | [ 0.66137183, 0.00869417, 0.89220798]], dtype=float32) |
| 1462 | ``` |
| 1463 | """ |
| 1464 | if dtype is None: |
| 1465 | dtype = floatx() |
| 1466 | tf_dtype = dtypes_module.as_dtype(dtype) |
| 1467 | if seed is None: |
| 1468 | # ensure that randomness is conditioned by the Numpy RNG |
| 1469 | seed = np.random.randint(10e8) |
| 1470 | value = init_ops.random_uniform_initializer( |
| 1471 | low, high, dtype=tf_dtype, seed=seed)(shape) |
| 1472 | return variable(value, dtype=dtype, name=name) |
| 1473 | |
| 1474 | |
| 1475 | @keras_export('keras.backend.random_normal_variable') |