custom op error #57742
Labels
comp:runtime
c++ runtime, performance issues (cpu)
stat:awaiting tensorflower
Status - Awaiting response from tensorflower
TF 2.9
Issues found in the TF 2.9 release (or RCs)
type:support
Support issues
To replace the tf.py_func, I wrote custom op by c++ and use tf.load_op_library to load the op in python.And get Segmentation fault (core dumped).And when I comment SetShapeFn, there is no error. But the output shape of the custom op is unknow.Could someone can help me ? Another info is using pb model to inference by c++ custom op is correct. Below is the error and code
######################################################################################################
This is the error.
######################################################################################################
WARNING:tensorflow:From` /home/zhj/Py/frustum-pointnets-master/models/tf_util.py:384: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.
WARNING:tensorflow:From /home/zhj/Py/frustum-pointnets-master/models/tf_util.py:613: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.
Instructions for updating:
Please use
rate
instead ofkeep_prob
. Rate should be set torate = 1 - keep_prob
.WARNING:tensorflow:From /home/zhj/Py/frustum-pointnets-master/models/model_util.py:235: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use
tf.cast
instead.WARNING:tensorflow:From /home/zhj/Py/frustum-pointnets-master/models/model_util.py:236: calling reduce_sum_v1 (from tensorflow.python.ops.math_ops) with keep_dims is deprecated and will be removed in a future version.
Instructions for updating:
keep_dims is deprecated, use keepdims instead
mask.shape---------------------------- (1, 1024)
Segmentation fault (core dumped)
######################################################################################################
This is the custom op
######################################################################################################
This is the python code
The text was updated successfully, but these errors were encountered: