Hi. I try to use nfnets, but I get the following error.

Error

AttributeError: 'NFNet' object has no attribute 'variant'

environment

• Ubuntu : 18.04
• Tensorflow : version 2.3 and 2.4 both give the same error.
• Keras : 2.4.3
• NFNets : 0.0.2

code

from nfnets_keras import NFNetF3, SGD_AGC

model = NFNetF3(include_top=True, num_classes=10)
model.compile( SGD_AGC(lr=1e-3), loss='categorical_crossentropy' )
model.summary()    

How can i solve this Error?

Hi!

I am trying to have a look at the TensorFlow implementation of NFNet, and somehow got to fix the reference-before-assignment issue with some help from @prateekkrjain.

However, when I try to instantiate NFNetF0 as follows: x = NFNetF0(num_classes = num_classes)(x)

the Python interpreter complains that it has an error as follows:

TypeError: in user code:

    C:\Users\<myUserName>\NFNet\nfnets_keras\nfnet.py:143 call  *
        for i, block in enumerate(self.blocks): out, res_avg_var = block(out, training = training)
    C:\Users\<myUserName>\NFNet\nfnets_keras\nfnet.py:200 call  *
        out = self.stoch_depth(out, training)
    C:\Users\<myUserName>\NFNet\nfnets_keras\nfnet_layers.py:47 call  *
        r = tf.random.uniform(shape = [batch_size, 1, 1, 1], dtype = x.dtype)
    C:\Users\<myUserName>\miniconda3\lib\site-packages\tensorflow\python\util\dispatch.py:201 wrapper  **
        return target(*args, **kwargs)
    C:\Users\<myUserName>\miniconda3\lib\site-packages\tensorflow\python\ops\random_ops.py:289 random_uniform
        shape = tensor_util.shape_tensor(shape)
    C:\Users\<myUserName>\miniconda3\lib\site-packages\tensorflow\python\framework\tensor_util.py:1035 shape_tensor
        return ops.convert_to_tensor(shape, dtype=dtype, name="shape")
    C:\Users\<myUserName>\miniconda3\lib\site-packages\tensorflow\python\profiler\trace.py:163 wrapped
        return func(*args, **kwargs)
    C:\Users\<myUserName>\miniconda3\lib\site-packages\tensorflow\python\framework\ops.py:1540 convert_to_tensor
        ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
    C:\Users\<myUserName>\miniconda3\lib\site-packages\tensorflow\python\framework\constant_op.py:339 _constant_tensor_conversion_function
        return constant(v, dtype=dtype, name=name)
    C:\Users\<myUserName>\miniconda3\lib\site-packages\tensorflow\python\framework\constant_op.py:264 constant
        return _constant_impl(value, dtype, shape, name, verify_shape=False,
    C:\Users\<myUserName>\miniconda3\lib\site-packages\tensorflow\python\framework\constant_op.py:281 _constant_impl
        tensor_util.make_tensor_proto(
    C:\Users\<myUserName>\miniconda3\lib\site-packages\tensorflow\python\framework\tensor_util.py:551 make_tensor_proto
        raise TypeError("Failed to convert object of type %s to Tensor. "

    TypeError: Failed to convert object of type <class 'tuple'> to Tensor. Contents: (None, 1, 1, 1). Consider casting elements to a supported type.

It seems like there is something wrong with the arguments being passed to tf.random.uniform() function when initializing StochasticDepth Layer. The batch_size turns out to be None.

Any ideas on how to fix this?

Note that I am using the latest version of Miniconda with Python 3.8, along with TensorFlow 2.4.1 on a Windows 10.

Hello Sir,

I am trying to use NFNET model, but I am encountering, the aforementioned error. Code:

Error:

AttributeError: 'NFNet' object has no attribute 'variant'

I believe The problem is in : https://github.com/ypeleg/nfnets-keras/blob/master/nfnets_keras/nfnet.py The above said line (line :: 89 block_params = self.variant_dict[self.variant]) should be after line 103(self.group_pattern = block_params.get('group_width', [128] * 4)). So the self.variant used in line 89 will be declared before it is used.

Thank you!

Hell Sir.

I tried your code but I met some error.

This is my code.

    optimizer = SGD_AGC(lr=0.01)

Error :

   TypeError: __init__() missing 1 required positional argument: 'params'

So I checked your code.

class SGD_AGC(Optimizer):
    ...
    def __init__(self, params, lr, ....):
    ...

I think that this is the style of optimizer of pytorch. How to use SGD_AGC class in keras.

Thanks..

thanks for the implementation of nfnets in keras!

I want to run it using optimizer(SGD_AGC). But there's an error like this.

TypeError: Can't instantiate abstract class SGD_AGC with abstract methods get_config