无法将 tensorflow.python.framework.ops.Tensor 对象转换为 numpy 数组以在 sklearn.metrics.cohen_kappa_score 函数中传递它

Question

我考虑使用 sklearn.metrics.cohen_kappa_score

实施 kappaScore 指标

def kappaScore(y_true,y_pred):
    k = cohen_kappa_score(y_true,y_pred,weights='quadratic')
    return k

当我尝试运行此代码时出现错误：

OperatorNotAllowedInGraphError: in user code:

    /opt/conda/lib/python3.7/site-packages/keras/engine/training.py:853 train_function  *
        return step_function(self, iterator)
    /tmp/ipykernel_33/1006337667.py:2 kappaScore  *
        k = cohen_kappa_score(y_true,y_pred,weights='quadratic')
    /opt/conda/lib/python3.7/site-packages/sklearn/utils/validation.py:555 inner_f  *
        return f(**kwargs)
    /opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py:600 cohen_kappa_score  *
        confusion = confusion_matrix(y1, y2, labels=labels,
    /opt/conda/lib/python3.7/site-packages/sklearn/utils/validation.py:555 inner_f  *
        return f(**kwargs)
    /opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py:276 confusion_matrix  *
        y_type, y_true, y_pred = _check_targets(y_true, y_pred)
    /opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py:81 _check_targets  *
        check_consistent_length(y_true, y_pred)
    /opt/conda/lib/python3.7/site-packages/sklearn/utils/validation.py:253 check_consistent_length  *
        uniques = np.unique(lengths)
    <__array_function__ internals>:6 unique  **
        
    /opt/conda/lib/python3.7/site-packages/numpy/lib/arraysetops.py:261 unique
        ret = _unique1d(ar, return_index, return_inverse, return_counts)
    /opt/conda/lib/python3.7/site-packages/numpy/lib/arraysetops.py:322 _unique1d
        ar.sort()
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/framework/ops.py:900 __bool__
        self._disallow_bool_casting()
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/framework/ops.py:504 _disallow_bool_casting
        "using a `tf.Tensor` as a Python `bool`")
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/framework/ops.py:491 _disallow_when_autograph_enabled
        " indicate you are trying to use an unsupported feature.".format(task))

    OperatorNotAllowedInGraphError: using a `tf.Tensor` as a Python `bool` is not allowed: AutoGraph did convert this function. This might indicate you are trying to use an unsupported feature.

这里y_true和y_pred的类型需要在列表或numpy数组中

但是y_true和y_pred的类型是，

y_true: <class 'tensorflow.python.framework.ops.Tensor'>
y_pred: <class 'tensorflow.python.framework.ops.Tensor'>

直接打印时（即不使用type()函数），显示如下：

y_true: Tensor("IteratorGetNext:1", shape=(None, None), dtype=float32)
y_pred: Tensor("sequential_5/dense_5/Softmax:0", shape=(None, 5), dtype=float32)

无法使用 y_true.numpy() () and tf.make_ndarray(y_true) (https://www.tensorflow.org/api_docs/python/tf/make_ndarray#:~:text=tf.make_ndarray(proto_tensor)) 试过了..

如何将这些数据类型转换为可以传递给 sklearn.metrics.cohen_kappa_score 函数的方式？ 我不想写kappa评分的代码。可以转换吗？

Answer 1

有一种方法可以解决将 cohen_kappa_score 包装在 tf.py_function 中的问题。它在tensorflow 2.x中可用，但我不知道从哪个版本的框架开始； py_function 为您完成所有繁重的工作，将 Python 函数包装到一个急切执行它的 TensorFlow 操作中。

import tensorflow as tf
import numpy as np
from sklearn.metrics import cohen_kappa_score


def cohen_kappa_score_wrapper(y_true, y_pred):
  y_pred = np.argmax(y_pred, axis=1)
  return cohen_kappa_score(y_true, y_pred, weights='quadratic')


def kappaScore(y_true, y_pred):
  return tf.py_function(
      func=cohen_kappa_score_wrapper,
      inp=[y_true, y_pred],
      Tout=tf.float32
    )


model.compile('adam', 'sparse_categorical_crossentropy', metrics=[kappaScore])

首先，定义cohen_kappa_score_wrapper。这很重要，因为最后一个密集层通常 returns 每个样本的每个 class 的概率数组。但是 Cohen kappa 分数接受 classes 的整数标签，因此必须将概率转换为带有 np.argmax() 的标签。我们仍在 Python 的领地内，所以可以只使用 numpy 函数。

然后用 py_function 包裹 cohen_kappa_score_wrapper：参见 kappaScore。

使用 MNIST 的完整示例：

import keras 
import tensorflow as tf
from keras.layers import Input, LSTM, RepeatVector, TimeDistributed, Dense
from keras import Model
import numpy as np

from sklearn.metrics import cohen_kappa_score
from keras.datasets import mnist

(x_train, x_test), (y_train, y_test) = mnist.load_data()
model = keras.Sequential([
   keras.layers.Flatten(),
   keras.layers.Dense(128, activation='relu'),
   keras.layers.Dense(10, activation='softmax'),
])

def cohen_kappa_score_wrapper(y_true, y_pred):
  y_pred = np.argmax(y_pred, axis=1)
  return cohen_kappa_score(y_true, y_pred, weights='quadratic')


def kappaScore(y_true, y_pred):
  return tf.py_function(
      func=cohen_kappa_score_wrapper,
      inp=[y_true, y_pred], #  weights='quadratic']
      Tout=tf.float32
    )

model.compile('adam', 'sparse_categorical_crossentropy', metrics=[kappaScore])
model.fit(x_train / 255., x_test)

60000/60000 [==============================] - 5s 90us/sample - loss: 0.2612 - kappaScore: 0.9202
<keras.callbacks.History at 0x7fcea289cc50>

备注尽管 docs 声明包装函数急切执行，但如果我关闭急切执行，它仍然对我有用：tf.compat.v1.disable_eager_execution()。我使用 tf2.7。但我对 framework/different 环境的其他版本不太有信心。有时这可能很棘手。此外，如果您使用 tf1.x，情况可能会有所不同。

无法将 tensorflow.python.framework.ops.Tensor 对象转换为 numpy 数组以在 sklearn.metrics.cohen_kappa_score 函数中传递它

Unable to convert tensorflow.python.framework.ops.Tensor object to numpy array for passoing it in sklearn.metrics.cohen_kappa_score function

python

numpy

scikit-learn

deep-learning

tensorflow