onnxruntime/python/_sources/auto_examples/plot_dl_keras.rst.txt
2019-12-20 13:35:58 -08:00

226 lines
4.3 KiB
ReStructuredText

.. note::
:class: sphx-glr-download-link-note
Click :ref:`here <sphx_glr_download_auto_examples_plot_dl_keras.py>` to download the full example code
.. rst-class:: sphx-glr-example-title
.. _sphx_glr_auto_examples_plot_dl_keras.py:
.. _l-example-backend-api-tensorflow:
ONNX Runtime for Keras
======================
The following demonstrates how to compute the predictions
of a pretrained deep learning model obtained from
`keras <https://keras.io/>`_
with *onnxruntime*. The conversion requires
`keras <https://keras.io/>`_,
`tensorflow <https://www.tensorflow.org/>`_,
`keras-onnx <https://github.com/onnx/keras-onnx/>`_,
`onnxmltools <https://pypi.org/project/onnxmltools/>`_
but then only *onnxruntime* is required
to compute the predictions.
.. code-block:: python
import os
if not os.path.exists('dense121.onnx'):
from keras.applications.densenet import DenseNet121
model = DenseNet121(include_top=True, weights='imagenet')
from keras2onnx import convert_keras
onx = convert_keras(model, 'dense121.onnx')
with open("dense121.onnx", "wb") as f:
f.write(onx.SerializeToString())
Let's load an image (source: wikipedia).
.. code-block:: python
from keras.preprocessing.image import array_to_img, img_to_array, load_img
img = load_img('Sannosawa1.jpg')
ximg = img_to_array(img)
import matplotlib.pyplot as plt
plt.imshow(ximg / 255)
plt.axis('off')
.. image:: /auto_examples/images/sphx_glr_plot_dl_keras_001.png
:class: sphx-glr-single-img
Let's load the model with onnxruntime.
.. code-block:: python
import onnxruntime as rt
from onnxruntime.capi.onnxruntime_pybind11_state import InvalidGraph
try:
sess = rt.InferenceSession('dense121.onnx')
ok = True
except (InvalidGraph, TypeError, RuntimeError) as e:
# Probably a mismatch between onnxruntime and onnx version.
print(e)
ok = False
if ok:
print("The model expects input shape:", sess.get_inputs()[0].shape)
print("image shape:", ximg.shape)
.. rst-class:: sphx-glr-script-out
Out:
.. code-block:: none
The model expects input shape: ['N', 224, 224, 3]
image shape: (960, 1280, 3)
Let's resize the image.
.. code-block:: python
if ok:
from skimage.transform import resize
import numpy
ximg224 = resize(ximg / 255, (224, 224, 3), anti_aliasing=True)
ximg = ximg224[numpy.newaxis, :, :, :]
ximg = ximg.astype(numpy.float32)
print("new shape:", ximg.shape)
.. rst-class:: sphx-glr-script-out
Out:
.. code-block:: none
new shape: (1, 224, 224, 3)
Let's compute the output.
.. code-block:: python
if ok:
input_name = sess.get_inputs()[0].name
res = sess.run(None, {input_name: ximg})
prob = res[0]
print(prob.ravel()[:10]) # Too big to be displayed.
.. rst-class:: sphx-glr-script-out
Out:
.. code-block:: none
[5.3862054e-06 6.1927537e-08 2.0104133e-07 1.3239808e-06 1.0415122e-06
6.7792627e-08 3.1348068e-07 3.2683040e-08 1.0914272e-07 1.1627122e-06]
Let's get more comprehensive results.
.. code-block:: python
if ok:
from keras.applications.densenet import decode_predictions
decoded = decode_predictions(prob)
import pandas
df = pandas.DataFrame(decoded[0], columns=["class_id", "name", "P"])
print(df)
.. rst-class:: sphx-glr-script-out
Out:
.. code-block:: none
class_id name P
0 n09468604 valley 0.785549
1 n09193705 alp 0.187054
2 n09399592 promontory 0.010602
3 n09246464 cliff 0.005537
4 n02417914 ibex 0.001997
**Total running time of the script:** ( 0 minutes 56.895 seconds)
.. _sphx_glr_download_auto_examples_plot_dl_keras.py:
.. only :: html
.. container:: sphx-glr-footer
:class: sphx-glr-footer-example
.. container:: sphx-glr-download
:download:`Download Python source code: plot_dl_keras.py <plot_dl_keras.py>`
.. container:: sphx-glr-download
:download:`Download Jupyter notebook: plot_dl_keras.ipynb <plot_dl_keras.ipynb>`
.. only:: html
.. rst-class:: sphx-glr-signature
`Gallery generated by Sphinx-Gallery <https://sphinx-gallery.readthedocs.io>`_