我正在尝试使用davidsandberg/facenet TF Lite Converter在Ubuntu 18.04.1 LTS(VirtualBox)上将我从(this is the specific model i am using)冻结的模型转换为.tflite。 当我尝试运行命令时:
/home/nils/.local/bin/tflite_convert
--output_file=/home/nils/Documents/frozen.tflite
--graph_def_file=/home/nils/Documents/20180402-114759/20180402-114759.pb
--input_arrays=input --output_array=embeddings
我收到以下错误:
2018-11-29 16:36:21.774098: I
tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports
instructions that this TensorFlow binary was not compiled to use: AVX2
Traceback (most recent call last):
File "/home/nils/.local/bin/tflite_convert", line 11, in <module>
sys.exit(main())
File
"/home/nils/.local/lib/python3.6/site-packages/tensorflow/contrib /lite/python/tflite_convert.py",
line 412, in main
app.run(main=run_main, argv=sys.argv[:1])
File
"/home/nils/.local/lib/python3.6/site-packages/tensorflow/python/platform/app.py",
line 125, in run
_sys.exit(main(argv))
File
"/home/nils/.local/lib/python3.6/site-packages/tensorflow/contrib/lite/python/tflite_convert.py",
line 408, in run_main
_convert_model(tflite_flags)
File
"/home/nils/.local/lib/python3.6/site-packages/tensorflow/contrib/lite/python/tflite_convert.py",
line 162, in _convert_model
output_data = converter.convert()
File
"/home/nils/.local/lib/python3.6/site-packages/tensorflow/contrib/lite/python/lite.py",
line 404, in convert
"'{0}'.".format(_tensor_name(tensor)))
ValueError: Provide an input shape for input array 'input'.
由于我自己尚未训练模型,因此我不知道输入的确切形状。可能有人可以从facenet / src的David Sandberg的GitHubRep中找到的classifier.py和facenet.py中提取它,但是我自己对代码的理解不足。 我什至尝试通过张量板分析该图。我仍然无法弄清楚,但也许你可以:Tensorboard-Screenshot 如您可能已经注意到的,我对Ubuntu,Tensorflow和所有相关的东西还很陌生,所以我很乐意就此问题提出任何建议。 预先谢谢你!
这是classifier.py的相关部分,在其中加载并设置模型:
# Load the model
print('Loading feature extraction model')
facenet.load_model(args.model)
# Get input and output tensors
images_placeholder = tf.get_default_graph().get_tensor_by_name("input:0")
embeddings = tf.get_default_graph().get_tensor_by_name("embeddings:0")
phase_train_placeholder = tf.get_default_graph().get_tensor_by_name("phase_train:0")
embedding_size = embeddings.get_shape()[1]
# Run forward pass to calculate embeddings
print('Calculating features for images')
nrof_images = len(paths)
nrof_batches_per_epoch = int(math.ceil(1.0*nrof_images / args.batch_size))
emb_array = np.zeros((nrof_images, embedding_size))
for i in range(nrof_batches_per_epoch):
start_index = i*args.batch_size
end_index = min((i+1)*args.batch_size, nrof_images)
paths_batch = paths[start_index:end_index]
images = facenet.load_data(paths_batch, False, False, args.image_size)
feed_dict = { images_placeholder:images, phase_train_placeholder:False }
emb_array[start_index:end_index,:] = sess.run(embeddings, feed_dict=feed_dict)
classifier_filename_exp = os.path.expanduser(args.classifier_filename)
答案 0 :(得分:0)
如果再次启动Tensorboard,请返回到已看到的图形,应该有一个搜索图标(我认为在左上角),您可以在其中输入“ input”并找到输入张量。它会为您提供所需的形状。我猜想它将是“ [1,image_size,image_size,3]”形式的东西。
或者,您可以检查代码
feed_dict = { images_placeholder:images, phase_train_placeholder:False }
请注意,我们正在将“ images”对象输入到images_placeholder中,该对象映射到“ input:0”张量。然后,您基本上需要图像对象的形状。
图像来自对facenet.load_data()的调用。如果进入facenet.py并检查load_data函数,可以观察到形状类似于我上面建议的形状。如果您打印image_size值,则该值应与您在Tensorboard中看到的值匹配。
答案 1 :(得分:0)
谢谢您的帮助,我确实像Alan Chiao所说,并按照load_data()到facenet.py,最终在这里找到了形状[1,160,160,3]。另外,Tensorflow's command line reference for the tf lite converter向我展示了我必须注意的事项:
-input_shapes。类型:用冒号分隔的列表,用逗号分隔的列表 整数每个用逗号分隔的整数列表都给出了 TensorFlow约定中指定的输入数组之一。
示例:对于典型的视觉模型,--input_shapes = 1,60,80,3表示批处理大小为1,输入图像高度为60,输入图像宽度为80,输入图像深度为3(表示RGB通道)。
答案 2 :(得分:0)
我经历了tflite转换器的代码。我发现您需要以{"input_tensor_name": [input shape]}
格式将输入形状作为字典。
以下是解决此问题的示例:
`graph_def_file = "20180402-114759/20180402-114759.pb"
input_arrays = ["input"]
output_arrays = ["embeddings"]
converter = tf.lite.TFLiteConverter.from_frozen_graph(
graph_def_file, input_arrays, output_arrays,input_shapes={"input":[1,160,160,3]})
tflite_model = converter.convert()
open("model.tflite", "wb").write(tflite_model)
`