Menu
Caffe can not only print overall accuracy, but also per-class accuracy.
In Keras log, there's only overall accuracy. It's hard for me to calculate the separate class accuracy.
Epoch 168/200 0s - loss: 0.0495 - acc: 0.9818 - val_loss: 0.0519 - val_acc: 0.9796 Epoch 169/200 0s - loss: 0.0519 - acc: 0.9796 - val_loss: 0.0496 - val_acc: 0.9815 Epoch 170/200 0s - loss: 0.0496 - acc: 0.9815 - val_loss: 0.0514 - val_acc: 0.9801 Anybody who knows how to output per-class accuracy in keras?
819
Accuracy(name="accuracy", dtype=None) Calculates how often predictions equal labels. This metric creates two local variables, total and count that are used to compute the frequency with which y_pred matches y_true .
metrics. accuracy and pass it to eval_metric_ops that will be returned by the function. Then the output of estimator. evaluate() will contain an accuracy key that will hold the accuracy computed on the validation set.
Accuracy is a metric that generally describes how the model performs across all classes. It is useful when all classes are of equal importance. It is calculated as the ratio between the number of correct predictions to the total number of predictions.
Precision & recall are more useful measures for multi-class classification (see definitions). Following the Keras MNIST CNN example (10-class classification), you can get the per-class measures using classification_report from sklearn.metrics:
from sklearn.metrics import classification_report import numpy as np Y_test = np.argmax(y_test, axis=1) # Convert one-hot to index y_pred = model.predict_classes(x_test) print(classification_report(Y_test, y_pred)) Here is the result:
precision recall f1-score support 0 0.99 1.00 1.00 980 1 0.99 0.99 0.99 1135 2 1.00 0.99 0.99 1032 3 0.99 0.99 0.99 1010 4 0.98 1.00 0.99 982 5 0.99 0.99 0.99 892 6 1.00 0.99 0.99 958 7 0.97 1.00 0.99 1028 8 0.99 0.99 0.99 974 9 0.99 0.98 0.99 1009 avg / total 0.99 0.99 0.99 10000 If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
