python - Python に基本的な tensorflow モデルがあり、それを onnx ファイルに変換したい

Question

Python tensorflow に基本モデルがあります。これを onnx ファイルに保存するにはどうすればよいですか。関数を使用しようとしonnx.saveましたが、エラーが発生しました。

File "tenserflowbase.py", line 21, in <module> onnx.save(trained_model,'model.onxx') 
File "C:\Users\Parag_IK\Anaconda3\lib\site-packages\onnx\__init__.py", line 184, in save_model proto = write_external_data_tensors(proto, basepath)                                                                                                                                                                                                                                                           
File "C:\Users\Parag_IK\Anaconda3\lib\site packages\onnx\external_data_helper.py", line 225, in write_external_data_tensors                                               
for tensor in _get_all_tensors(model):                                                                                                                              
File "C:\Users\Parag_IK\Anaconda3\lib\site packages\onnx\external_data_helper.py", line 170, in _get_initializer_tensors                                                 
for initializer in onnx_model_proto.graph.initializer:                                                                                                              
AttributeError: 'History' object has no attribute 'graph'** 

私のコードは以下の通りです：

import onnx
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split

tf.logging.set_verbosity(tf.logging.ERROR)   

mar_budget = np.array([60, 80,  100  , 30, 50, 20, 90,  10],  dtype=float)
subs_gained = np.array([160, 200, 240, 100, 140, 80, 220, 60],  dtype=float)

for i, c in enumerate(mar_budget):
  print("{} Market budget = {} new subscribers gained".format(c, subs_gained[i]))


X_train, X_test, y_train, y_test = train_test_split(mar_budget, subs_gained, 
   random_state=42, train_size=0.8, test_size=0.2)

layer_0 = tf.keras.layers.Dense(units=1, input_shape=[1])
model = tf.keras.Sequential([layer_0])
model = tf.keras.Sequential([layer_0])

model.compile(loss='mean_squared_error', optimizer=tf.keras.optimizers.Adam(0.1))

trained_model = model.fit(X_train, y_train, epochs=1000, verbose=False)      

onnx.save(trained_model,'model.onxx')

print("Finished training the model")   
print(model.predict([80.0]))