What are the differences between the mean squared error function in tf.keras.metrics.mean_squared_error and tf.keras.losses.MeanSquaredError?
Asked
Active
Viewed 1,252 times
0
-
Refer to [this](https://stackoverflow.com/questions/48280873/what-is-the-difference-between-loss-function-and-metric-in-keras) question. Mean squared error can be used as a loss function as well as a metric. – Shubham Panchal Sep 27 '21 at 02:27
1 Answers
1
The former is used as an indicator, and not used in the backpropagation calculation for updating the weights. It is used if you use other function as a loss function, but at the same time you also want to know the MSE value.

sgd
- 136
- 3