版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/sinat_27935693/article/details/81015398
当出现TypeError: Fetch argument 7.2057328 has invalid type <class 'numpy.float32'>, must be a string or Tensor. (Can not convert a float32 into a Tensor or Operation.)
这种错误时，修改如下：
将
loss = sess.run(loss, feed_dict={x: batch_tx, y: batch_ty, dropout_keep_prob:1.} )修改为
los = sess.run(loss, feed_dict={x: batch_tx, y: batch_ty, dropout_keep_prob:1.} )