In [1]:

    

input_fld = '../data'
weight_file = 'weights.best.inceptionv3_finetune.hdf5'
num_output = 1
write_graph_def_ascii_flag = True
prefix_output_node_names_of_final_network = 'softmax'


    

In [2]:

    

from keras.models import load_model
import tensorflow as tf
import os
import os.path as osp
from keras import backend as K

weight_file_path = osp.join(input_fld, weight_file)


    

    




Using TensorFlow backend.

In [3]:

    

net_model = load_model(weight_file_path)
K.set_learning_phase(0)

pred = [None]*num_output
pred_node_names = [None]*num_output
for i in range(num_output):
    pred_node_names[i] = prefix_output_node_names_of_final_network+str(i)
    pred[i] = tf.identity(net_model.output[i], name=pred_node_names[i])
print('output nodes names are: ', pred_node_names)


    

    




output nodes names are:  ['softmax0']

In [4]:

    

sess = K.get_session()


    

In [5]:

    

gd = sess.graph.as_graph_def()
nodes_with_switch_op = [x for x in gd.node if x.op.lower().find('switch') != -1]


    

In [6]:

    

# nodes to reroute switch input
for n in [x for x in gd.node]:
    ints = [i for i in n.input]
    endswith1 = [ii for ii in ints if ii.endswith(':1')] #and 'Switch' in ii]
    if len(endswith1) > 0:
        for index, inn in enumerate(n.input):
            if inn in endswith1:
                new_input = inn[:-2]
                n.input.remove(inn)
                n.input.insert(index, new_input)


    

In [7]:

    

nodes = {}
for node in gd.node:
    nodes[node.name] = node


    

In [8]:

    

for n in nodes_with_switch_op:
    n.op = 'Identity'
    n.input.pop()


    

In [9]:

    

nodes_with_switch_op = [x for x in gd.node if x.op.lower().find('switch') != -1]


    

In [10]:

    

[(n.name, [i for i in n.input]) for n in gd.node if n.name.lower().find('keras_learning_phase') != -1]


    

    
Out[10]:





[('batch_normalization_95/keras_learning_phase', [])]

In [11]:

    

nodes['batch_normalization_95/keras_learning_phase']


    

    
Out[11]:





name: "batch_normalization_95/keras_learning_phase"
op: "Placeholder"
attr {
  key: "dtype"
  value {
    type: DT_BOOL
  }
}
attr {
  key: "shape"
  value {
    shape {
      unknown_rank: true
    }
  }
}

In [12]:

    

nodes['batch_normalization_95/keras_learning_phase'].op = 'Const'


    

In [13]:

    

from tensorflow.core.framework import attr_value_pb2
from tensorflow.core.framework import graph_pb2
from tensorflow.python.framework import dtypes
from tensorflow.python.framework import tensor_util
nodes['batch_normalization_95/keras_learning_phase'].attr.get_or_create('value').CopyFrom(
    attr_value_pb2.AttrValue(tensor=tensor_util.make_tensor_proto([False], dtypes.bool, [1])))


    

In [14]:

    

del nodes['batch_normalization_95/keras_learning_phase'].attr['shape']


    

In [15]:

    

nodes['batch_normalization_95/keras_learning_phase']


    

    
Out[15]:





name: "batch_normalization_95/keras_learning_phase"
op: "Const"
attr {
  key: "dtype"
  value {
    type: DT_BOOL
  }
}
attr {
  key: "value"
  value {
    tensor {
      dtype: DT_BOOL
      tensor_shape {
        dim {
          size: 1
        }
      }
      bool_val: false
    }
  }
}

In [16]:

    

from tensorflow.python.framework import graph_util

output_graph_def = graph_util.convert_variables_to_constants(sess, gd, pred_node_names)

with tf.gfile.GFile('../data/dog_breed_graph_v2.pb', "wb") as f:
    f.write(output_graph_def.SerializeToString())


    

    




INFO:tensorflow:Froze 380 variables.
Converted 380 variables to const ops.

In [ ]: