Feature: Flops calculation

[innat]

Reopening from

• Feature: Flops calculation #138
• TF 2.x Feature: Flops calculation  #552
• TF 2.0 Feature: Flops calculation tensorflow/tensorflow#32809
• Flops count? keras#12970
• Report number of floating point operations in model summary keras#5625
• Flops calculation in tensorflow 2.x version. tensorflow/tensorflow#39834

Currently the adopted solutions (from community) using tf.

import tensorflow as tf
from tensorflow.python.profiler import model_analyzer, option_builder

model = tf.keras.applications.Xception(
    weights='imagenet',
    input_shape=(150, 150, 3),
    include_top=False
) 

input_signature = [
    tf.TensorSpec(
        shape=(1, *params.shape[1:]), 
        dtype=params.dtype, 
        name=params.name
    ) for params in model.inputs
]
forward_graph = tf.function(model, input_signature).get_concrete_function().graph
options = option_builder.ProfileOptionBuilder.float_operation()
graph_info = model_analyzer.profile(forward_graph, options=options)
flops = graph_info.total_float_ops // 2
flops # 1925897756 

And using official facebookresearch,

• https://github.com/facebookresearch/fvcore