Get the tensor of type `dtype` by feeding a tensor handle. This is EXPERIMENTAL and subject to change. Get the value of the tensor from a tensor handle. The tensor is produced in a previous run() and stored in the state of the session. Args: handle: The string representation of a pe
(handle, dtype, name=None)
| 181 | |
| 182 | @tf_export(v1=["get_session_tensor"]) |
| 183 | def get_session_tensor(handle, dtype, name=None): |
| 184 | """Get the tensor of type `dtype` by feeding a tensor handle. |
| 185 | |
| 186 | This is EXPERIMENTAL and subject to change. |
| 187 | |
| 188 | Get the value of the tensor from a tensor handle. The tensor |
| 189 | is produced in a previous run() and stored in the state of the |
| 190 | session. |
| 191 | |
| 192 | Args: |
| 193 | handle: The string representation of a persistent tensor handle. |
| 194 | dtype: The type of the output tensor. |
| 195 | name: Optional name prefix for the return tensor. |
| 196 | |
| 197 | Returns: |
| 198 | A pair of tensors. The first is a placeholder for feeding a |
| 199 | tensor handle and the second is the tensor in the session state |
| 200 | keyed by the tensor handle. |
| 201 | |
| 202 | Example: |
| 203 | |
| 204 | ```python |
| 205 | c = tf.multiply(a, b) |
| 206 | h = tf.compat.v1.get_session_handle(c) |
| 207 | h = sess.run(h) |
| 208 | |
| 209 | p, a = tf.compat.v1.get_session_tensor(h.handle, tf.float32) |
| 210 | b = tf.multiply(a, 10) |
| 211 | c = sess.run(b, feed_dict={p: h.handle}) |
| 212 | ``` |
| 213 | |
| 214 | """ |
| 215 | handle_device = TensorHandle._get_device_name(handle) |
| 216 | with ops.device(handle_device): |
| 217 | holder = array_ops.placeholder(dtypes.string) |
| 218 | _register_handle_feeder(holder.graph, holder, dtype) |
| 219 | tensor = gen_data_flow_ops.get_session_tensor(holder, dtype, name=name) |
| 220 | return (holder, tensor) |
| 221 | |
| 222 | |
| 223 | @tf_export(v1=["delete_session_tensor"]) |
nothing calls this directly
no test coverage detected