CODEWITHSUNDEEP
pythontensorflow
Lucas Pelizzari
Lucas Pelizzari

Reputation: 119

How to suppress "WARNING:tensorflow:AutoGraph could not transform <bound method Layer.__call__ of ... >>"?

I get this message every time I run tf.keras.Sequential().predict_on_batch:

WARNING: AutoGraph could not transform <bound method Layer.__call__ of <tensorflow.python.keras.engine.sequential.Sequential object at 0x000001F927581348>> and will run it as-is.
Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.
Cause:

It completely fills my console, and I can't find a way to suppress this.

Python version: 3.7.7 Tensorflow version: 2.1.0

Upvotes: 1

Views: 4902

Answers (2)

Stefan Meili
Stefan Meili

Reputation: 99

I've come across this in custom layers with for loops in them. I've been decorating the call method with @tf.autograph.experimental.do_not_convert

For example:

class DoNothing(tf.keras.layers.Layer):
    def __init__(self, **kwargs):
        super().__init__(**kwargs)
        
    def get_config(self):
        config = {}
        base_config = super().get_config()
        return {**base_config, **config}

    @tf.autograph.experimental.do_not_convert
    def call(self, args, **kwargs):
       return args # Yes, this layer really doesn't do anything

Upvotes: 2

user11530462
user11530462

Reputation:

Set the autograph verbosity level using the below code

import os
import tensorflow as tf
os.environ['AUTOGRAPH_VERBOSITY'] = 1

OR

You can just suppress warning by setting log level at the start of the program

import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2' 
import tensorflow as tf

Upvotes: 1

Related Questions