KERAS：InvalidArgumentError：无法计算 __inference_pruned_16232

问题描述

使用 Google 的 colab 我已经训练了一个 keras 模型。

Model: "sequential_1"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_4 (Dense)              (None,256)               131328    
_________________________________________________________________
dense_5 (Dense)              (None,128)               32896     
_________________________________________________________________
dense_6 (Dense)              (None,64)                8256      
_________________________________________________________________
dense_7 (Dense)              (None,22)                1430      
=================================================================
Total params: 173,910
Trainable params: 173,910
Non-trainable params: 0
_________________________________________________________________
None

但是，当我为我的测试数据尝试 model.predict() 时。这给了我以下错误。

InvalidArgumentError                      Traceback (most recent call last)
<ipython-input-13-2dd0bff687f9> in <module>()
      1 X_test_tensor = tf.convert_to_tensor(X_test)
----> 2 test_predict = model.predict(X_test)

6 frames
/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name,num_outputs,inputs,attrs,ctx,name)
     58     ctx.ensure_initialized()
     59     tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle,device_name,op_name,---> 60                                         inputs,num_outputs)
     61   except core._NotOkStatusException as e:
     62     if name is not None:

InvalidArgumentError: cannot compute __inference_pruned_16232 as input #0(zero-based) was expected to be a string tensor but is a float tensor [Op:__inference_pruned_16232]

解决方法

暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！

如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。

小编邮箱:dio#foxmail.com (将#修改为@）