python - 如何使用预训练的tensorflowlite模型进行输入解释

Question

我正在使用预先训练的 tensorflow 光模型来使用我创建的短程序获取一些示例输出:

import cv2 as cv
import numpy as np
import os
import tensorflow as tf
import numpy as np

def decode_img(img):
  # convert the compressed string to a 3D uint8 tensor
  img = tf.image.decode_jpeg(img, channels=3)
  # Use `convert_image_dtype` to convert to floats in the [0,1] range.
  img = tf.image.convert_image_dtype(img, tf.float32)
  # resize the image to the desired size.
  return tf.reshape(tf.image.resize(img, [257, 257]), [1, 257, 257, 3])


model = tf.lite.Interpreter('models\posenet_mobilenet_v1_100_257x257_multi_kpt_stripped.tflite')
input_details = model.get_input_details()
output_details = model.get_output_details()

img = tf.io.read_file('photos\standing\\1.jpg')
input_data = decode_img(img)

print('input shape: {}'.format(input_data.shape))
model.set_tensor(input_details[0]['index'], input_data)
model.invoke()

output_data = model.get_tensor(output_details[0]['index'])
print('output: {}'.format(output_data))

输入形状打印到控制台后，没有发生任何其他事情，程序结束。应该打印输出的行永远不会执行。

输出:

C:\python imagetest.py INFO: Initialized TensorFlow Lite runtime. 2020-01-21 08:07:32.567619: I tensorflow/core/platform/cpu_feature_guard.cc:145] This TensorFlow binary is optimized with Intel(R) MKL-DNN to use the following CPU instructions in performance critical operations: AVX AVX2 To enable them in non-MKL-DNN operations, rebuild TensorFlow with the appropriate compiler flags. 2020-01-21 08:07:32.578283: I tensorflow/core/common_runtime/process_util.cc:115] Creating new thread pool with default inter op setting: 8. Tune using inter_op_parallelism_threads for best performance. input shape: (1, 257, 257, 3)

通过 Interpreter 类使用预训练 tflite 模型的正确方法是什么？

Answer 1

能够使用以下代码使其正常工作:

img = cv.imread('photos\standing\\3.jpg')
img = tf.reshape(tf.image.resize(img, [257,257]), [1,257,257,3])
model = tf.lite.Interpreter('models\posenet_mobilenet_v1_100_257x257_multi_kpt_stripped.tflite')
model.allocate_tensors()
input_details = model.get_input_details()
output_details = model.get_output_details()
floating_model = input_details[0]['dtype'] == np.float32

if floating_model:
    img = (np.float32(img) - 127.5) / 127.5

model.set_tensor(input_details[0]['index'], img)
model.invoke()

output_data =  model.get_tensor(output_details[0]['index'])
offset_data = model.get_tensor(output_details[1]['index'])

我注意到的唯一区别是对 allocate_tensors() 的初始调用。