Instantiates an all-zeros variable and returns it. Arguments: shape: Tuple or list of integers, shape of returned Keras variable dtype: data type of returned Keras variable name: name of returned Keras variable Returns: A variable (including Keras metadata), filled with
(shape, dtype=None, name=None)
| 1267 | |
| 1268 | @keras_export('keras.backend.zeros') |
| 1269 | def zeros(shape, dtype=None, name=None): |
| 1270 | """Instantiates an all-zeros variable and returns it. |
| 1271 | |
| 1272 | Arguments: |
| 1273 | shape: Tuple or list of integers, shape of returned Keras variable |
| 1274 | dtype: data type of returned Keras variable |
| 1275 | name: name of returned Keras variable |
| 1276 | |
| 1277 | Returns: |
| 1278 | A variable (including Keras metadata), filled with `0.0`. |
| 1279 | Note that if `shape` was symbolic, we cannot return a variable, |
| 1280 | and will return a dynamically-shaped tensor instead. |
| 1281 | |
| 1282 | Example: |
| 1283 | |
| 1284 | ```python |
| 1285 | from tensorflow.keras import backend as K |
| 1286 | kvar = K.zeros((3,4)) |
| 1287 | K.eval(kvar) |
| 1288 | # array([[ 0., 0., 0., 0.], [ 0., 0., 0., 0.], |
| 1289 | # [ 0., 0., 0., 0.]], dtype=float32) |
| 1290 | A = tf.constant([1,2,3]) |
| 1291 | kvar2 = K.zeros(A.shape) # [0., 0., 0.] float32 by default |
| 1292 | kvar3 = K.zeros(A.shape,dtype=tf.int32) # [0, 0, 0] with int32 dtype |
| 1293 | kvar4 = K.zeros([2,3]) # [[0., 0., 0.], [0., 0., 0.]] |
| 1294 | ``` |
| 1295 | |
| 1296 | """ |
| 1297 | with ops.init_scope(): |
| 1298 | if dtype is None: |
| 1299 | dtype = floatx() |
| 1300 | tf_dtype = dtypes_module.as_dtype(dtype) |
| 1301 | v = array_ops.zeros(shape=shape, dtype=tf_dtype, name=name) |
| 1302 | if py_all(v.shape.as_list()): |
| 1303 | return variable(v, dtype=dtype, name=name) |
| 1304 | track_variable(v) |
| 1305 | return v |
| 1306 | |
| 1307 | |
| 1308 | @keras_export('keras.backend.ones') |
nothing calls this directly
no test coverage detected