Hi,
I am trying to convert Inception_v4.pb into inception_v4.plan. But there are many problems happened. Here is the process:
First step：
Fine-tune a inception_v4 model using transfer learning(without change structure, just train the classifier),and then I got a pb file named Inception_v4.pb. and I use Inception_v4.pb to test my testing dataset, and it can classify the pictures correctly.
Second step：
Type a command to convert this pb file into the plan file.
python scripts/convert_plan.py data/frozen_graphs/inception_v4.pb data/plans/inception_v4.plan input 299 299 InceptionV4/Logits/Logits/BiasAdd 1 0 float
But there are lots of errors, and I do not why.
The results of executing the command is as followed:
[b]Using output node InceptionV4/Logits/Logits/BiasAdd
Converting to UFF graph
DEBUG: convert reshape to flatten node
Warning: No conversion function registered for layer: Concat yet.
Converting as custom op Concat InceptionV4/InceptionV4/Mixed_7d/concat
name: “InceptionV4/InceptionV4/Mixed_7d/concat”
op: “Concat”
input: “InceptionV4/InceptionV4/Mixed_7d/concat/concat_dim”
input: “InceptionV4/InceptionV4/Mixed_7d/Branch_0/Conv2d_0a_1x1/Relu”
input: “InceptionV4/InceptionV4/Mixed_7d/Branch_1/concat”
input: “InceptionV4/InceptionV4/Mixed_7d/Branch_2/concat”
input: “InceptionV4/InceptionV4/Mixed_7d/Branch_3/Conv2d_0b_1x1/Relu”
attr {
key: “N”
value {
i: 4
}
}
attr {
key: “T”
value {
type: DT_FLOAT
}
}
Traceback (most recent call last):
File “scripts/convert_plan.py”, line 71, in
data_type
File “scripts/convert_plan.py”, line 22, in frozenToPlan
text=False,
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/conversion_helpers.py”, line 103, in from_tensorflow_frozen_model
return from_tensorflow(graphdef, output_nodes, **kwargs)
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/conversion_helpers.py”, line 75, in from_tensorflow
name=“main”)
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py”, line 64, in convert_tf2uff_graph
uff_graph, input_replacements)
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py”, line 51, in convert_tf2uff_node
op, name, tf_node, inputs, uff_graph, tf_nodes=tf_nodes)
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py”, line 28, in convert_layer
fields = cls.parse_tf_attrs(tf_node.attr)
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py”, line 177, in parse_tf_attrs
for key, val in attrs.items()}
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py”, line 177, in
for key, val in attrs.items()}
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py”, line 172, in parse_tf_attr_value
return cls.convert_tf2uff_field(code, val)
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py”, line 146, in convert_tf2uff_field
return TensorFlowToUFFConverter.convert_tf2numpy_dtype(val)
File “/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py”, line 74, in convert_tf2numpy_dtype
return np.dtype(dt[dtype])
TypeError: list indices must be integers, not AttrValue
[/b]
Please give me advises that what’s wrong with the code.
My enviroemnt is as followed:
- TensorRt 3.0
- tensorflow 1.5
Besides, I did some atttempts:
- Convert a official model named Inception_v4.ckpt into Inception_v4.pb ----> OK
- Convert acquired Inception_v4.pb into Inception_v4.plan → OK