TensorBoard 中的 Tensorflow 混淆矩阵 [英] Tensorflow Confusion Matrix in TensorBoard

问题描述

我想在张量板中有一个混淆矩阵的视觉效果.为此，我正在修改 Tensorflow Slim 的评估示例:

这里的功能几乎可以为您做所有事情.

from textwrap import wrap进口重新导入迭代工具导入 tfplot导入 matplotlib将 numpy 导入为 np从 sklearn.metrics 导入混淆_矩阵def plot_confusion_matrix(correct_labels, predict_labels, labels, title='混淆矩阵', tensor_name = 'MyFigure/image', normalize=False):'''参数:right_labels :这些是您真正的分类类别.predict_labels :这些是您预测的分类类别标签:这是一列标签，用于显示 axix 标签title='混淆矩阵':矩阵的标题tensor_name = 'MyFigure/image' : 输出 summay 张量的名称返回:总结:TensorFlow总结其他注意事项:- 根据类别和数据的数量，您可能需要修改 figzie、字体大小等.- 目前，由于轮换，一些刻度线不对齐.'''cm = 混淆矩阵(正确标签，预测标签，标签=标签)如果标准化:cm = cm.astype('float')*10/cm.sum(axis=1)[:, np.newaxis]cm = np.nan_to_num(cm, copy=True)cm = cm.astype('int')np.set_printoptions(precision=2)###fig, ax = matplotlib.figure.Figure()fig = matplotlib.figure.Figure(figsize=(7, 7), dpi=320, facecolor='w', edgecolor='k')ax = fig.add_subplot(1, 1, 1)im = ax.imshow(cm, cmap='Oranges')classes = [re.sub(r'([a-z](?=[A-Z])|[A-Z](?=[A-Z][a-z]))', r'1 ', x) 标签中的 x]classes = ['
'.join(wrap(l, 40)) for l in classes]tick_marks = np.arange(len(classes))ax.set_xlabel('预测', fontsize=7)ax.set_xticks(tick_marks)c = ax.set_xticklabels(classes, fontsize=4, rotation=-90, ha='center')ax.xaxis.set_label_position('底部')ax.xaxis.tick_bottom()ax.set_ylabel('真实标签', fontsize=7)ax.set_yticks(tick_marks)ax.set_yticklabels(classes, fontsize=4, va ='center')ax.yaxis.set_label_position('left')ax.yaxis.tick_left()对于 i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):ax.text(j, i, format(cm[i, j], 'd') if cm[i,j]!=0 else '.', horizo​​ntalalignment="center", fontsize=6, verticalalignment='center', color="黑色")fig.set_tight_layout(真)摘要 = tfplot.figure.to_summary(fig, tag=tensor_name)退货摘要

#

这是调用此函数所需的其余代码.

'''混淆矩阵总结'''img_d_summary_dir = os.path.join(checkpoint_dir, "summaries", "img")img_d_summary_writer = tf.summary.FileWriter(img_d_summary_dir, sess.graph)img_d_summary = plot_confusion_matrix(correct_labels, predict_labels, 标签, tensor_name='dev/cm')img_d_summary_writer.add_summary(img_d_summary, current_step)

迷惑！

I want to have a visual of confusion matrix in tensorboard. To do this, I am modifying Evaluation example of Tensorflow Slim: https://github.com/tensorflow/models/blob/master/slim/eval_image_classifier.py

In this example code, Accuracy already provided but it is not possible to add "confusion matrix" metric directly because it is not streaming.

What is difference between streaming metrics and non-streaming ones?

Therefore, I tried to add it like this:

c_matrix = slim.metrics.confusion_matrix(predictions, labels)

#These operations needed for image summary
c_matrix = tf.cast(c_matrix, uint8)
c_matrix = tf.expand_dims(c_matrix, 2)
c_matrix = tf.expand_dims(c_matrix, 0)

op = tf.image_summary("confusion matrix", c_matrix, collections=[])
tf.add_to_collection(tf.GraphKeys.SUMMARIES, op)

This creates an image in tensorboard but probably there is a formatting problem. Matrix should be normalized between 0-1 so that It produces meaningful image.

How can I produce a meaningful confusion matrix? How can I deal with multi batch evaluation process?

解决方案

Here is something I have put together That works reasonably well. Still need to adjust a few things like the tick placements etc.

Here is the function that will pretty much do everything for you.

from textwrap import wrap
import re
import itertools
import tfplot
import matplotlib
import numpy as np
from sklearn.metrics import confusion_matrix



def plot_confusion_matrix(correct_labels, predict_labels, labels, title='Confusion matrix', tensor_name = 'MyFigure/image', normalize=False):
''' 
Parameters:
    correct_labels                  : These are your true classification categories.
    predict_labels                  : These are you predicted classification categories
    labels                          : This is a lit of labels which will be used to display the axix labels
    title='Confusion matrix'        : Title for your matrix
    tensor_name = 'MyFigure/image'  : Name for the output summay tensor

Returns:
    summary: TensorFlow summary 

Other itema to note:
    - Depending on the number of category and the data , you may have to modify the figzie, font sizes etc. 
    - Currently, some of the ticks dont line up due to rotations.
'''
cm = confusion_matrix(correct_labels, predict_labels, labels=labels)
if normalize:
    cm = cm.astype('float')*10 / cm.sum(axis=1)[:, np.newaxis]
    cm = np.nan_to_num(cm, copy=True)
    cm = cm.astype('int')

np.set_printoptions(precision=2)
###fig, ax = matplotlib.figure.Figure()

fig = matplotlib.figure.Figure(figsize=(7, 7), dpi=320, facecolor='w', edgecolor='k')
ax = fig.add_subplot(1, 1, 1)
im = ax.imshow(cm, cmap='Oranges')

classes = [re.sub(r'([a-z](?=[A-Z])|[A-Z](?=[A-Z][a-z]))', r'1 ', x) for x in labels]
classes = ['
'.join(wrap(l, 40)) for l in classes]

tick_marks = np.arange(len(classes))

ax.set_xlabel('Predicted', fontsize=7)
ax.set_xticks(tick_marks)
c = ax.set_xticklabels(classes, fontsize=4, rotation=-90,  ha='center')
ax.xaxis.set_label_position('bottom')
ax.xaxis.tick_bottom()

ax.set_ylabel('True Label', fontsize=7)
ax.set_yticks(tick_marks)
ax.set_yticklabels(classes, fontsize=4, va ='center')
ax.yaxis.set_label_position('left')
ax.yaxis.tick_left()

for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
    ax.text(j, i, format(cm[i, j], 'd') if cm[i,j]!=0 else '.', horizontalalignment="center", fontsize=6, verticalalignment='center', color= "black")
fig.set_tight_layout(True)
summary = tfplot.figure.to_summary(fig, tag=tensor_name)
return summary

#

And here is the rest of the code that you will need to call this functions.

''' confusion matrix summaries '''
img_d_summary_dir = os.path.join(checkpoint_dir, "summaries", "img")
img_d_summary_writer = tf.summary.FileWriter(img_d_summary_dir, sess.graph)
img_d_summary = plot_confusion_matrix(correct_labels, predict_labels, labels, tensor_name='dev/cm')
img_d_summary_writer.add_summary(img_d_summary, current_step)

Confuse away!!!

这篇关于TensorBoard 中的 Tensorflow 混淆矩阵的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！