У меня есть .tflite, затем я хочу вывести веса каждого слоя, но он не может работать, он не может выводить все данные, но когда я вывожу имя слоя, он может выводить все имя. Я не знаю, как это исправить.
MyCode
import tensorflow as tf
interpreter = tf.lite.Interpreter(model_path='/content/drive/MyDrive/non_RR/tflite_qaware_model.tflite')
interpreter.allocate_tensors()
input_details = interpreter.get_tensor_details()
for i in range(0, 50):
print(input_details['name'])
print(interpreter.tensor(input_details['index'])(),'\n')
Вывод
serving_default_conv1d_input:0
[[[ -64]
[ 3]
[-124]
[ 123]
[ 2]
[ 124]
[ 0]
[ 0]
[ -64]
[ 3]
[-124]
[ 123]
[ 2]
[ 124]
[ 0]
[ 0]
[ 48]
[ 0]
[ 46]
[ 117]
[ -41]
[ 88]
[ 0]
[ 0]
[ 48]
[ 0]
[ 46]
[ 117]
[ -41]
[ 88]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]
[ 0]]]
sequential/quant_conv1d/Conv1D/ExpandDims/dim
-3
sequential/quant_conv1d/Conv1D/Squeeze
[ -1 256 4]
sequential/quant_average_pooling1d/ExpandDims/dim
2
sequential/quant_max_pooling1d/Squeeze
[ -1 128 4]
sequential/quant_conv1d_1/Conv1D/Squeeze
[ -1 128 16]
...
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]]]]
sequential/quant_conv1d/Relu;sequential/quant_conv1d/BiasAdd;sequential/quant_conv1d/Conv1D/Squeeze;sequential/quant_conv1d/BiasAdd/ReadVariableOp;sequential/quant_conv1d/Conv1D
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
in ()
6 for i in range(0, 50):#output_details[0]['name'])
7 print(input_details['name'])
----> 8 print(interpreter.tensor(input_details['index'])(),'\n')
/usr/local/lib/python3.10/dist-packages/tensorflow/lite/python/interpreter.py in ()
924 but it is not safe to hold the numpy array forever.
925 """
--> 926 return lambda: self._interpreter.tensor(self._interpreter, tensor_index)
927
928 def invoke(self):
ValueError: Tensor data is null. Run allocate_tensors() first
Подробнее здесь: https://stackoverflow.com/questions/781 ... ta-is-null
