dnn_backend_native_layer_mathunary: add atan support
authorTing Fu <ting.fu@intel.com>
Thu, 18 Jun 2020 09:15:35 +0000 (17:15 +0800)
committerGuo, Yejun <yejun.guo@intel.com>
Thu, 25 Jun 2020 00:41:50 +0000 (08:41 +0800)
commit13f5613e684f65e625cb287f85c3410f63efbf1c
tree0ee21fd3f50facd3995b2497655b618c990a4aa5
parent130c6001443c4de21203cb1f62e378042f0d51f8
dnn_backend_native_layer_mathunary: add atan support

It can be tested with the model generated with below python script:

import tensorflow as tf
import numpy as np
import imageio

in_img = imageio.imread('input.jpeg')
in_img = in_img.astype(np.float32)/255.0
in_data = in_img[np.newaxis, :]

x = tf.placeholder(tf.float32, shape=[1, None, None, 3], name='dnn_in')
x1 = tf.atan(x)
x2 = tf.divide(x1, 3.1416/4) # pi/4
y = tf.identity(x2, name='dnn_out')

sess=tf.Session()
sess.run(tf.global_variables_initializer())

graph_def = tf.graph_util.convert_variables_to_constants(sess, sess.graph_def, ['dnn_out'])
tf.train.write_graph(graph_def, '.', 'image_process.pb', as_text=False)

print("image_process.pb generated, please use \
path_to_ffmpeg/tools/python/convert.py to generate image_process.model\n")

output = sess.run(y, feed_dict={x: in_data})
imageio.imsave("out.jpg", np.squeeze(output))

Signed-off-by: Ting Fu <ting.fu@intel.com>
Signed-off-by: Guo Yejun <yejun.guo@intel.com>
libavfilter/dnn/dnn_backend_native_layer_mathunary.c
libavfilter/dnn/dnn_backend_native_layer_mathunary.h
tools/python/convert_from_tensorflow.py
tools/python/convert_header.py