The loss function is used to optimize your model. This is the function that will get minimized by the optimizer.
A metric is used to judge the performance of your model. This is only for you to look at and has nothing to do with the optimization process.
More Related Contents:
- Why binary_crossentropy and categorical_crossentropy give different performances for the same problem?
- Many to one and many to many LSTM examples in Keras
- How to calculate optimal batch size
- Keras Text Preprocessing – Saving Tokenizer object to file for scoring
- Is deep learning bad at fitting simple non linear functions outside training scope (extrapolating)?
- What is the difference between Keras model.evaluate() and model.predict()?
- Loss & accuracy – Are these reasonable learning curves?
- Common causes of nans during training of neural networks
- Keras Sequential model input layer
- How to interpret caffe log with debug_info?
- Caffe | solver.prototxt values setting strategy
- Epoch vs Iteration when training neural networks [closed]
- What is `weight_decay` meta parameter in Caffe?
- ResNet: 100% accuracy during training, but 33% prediction accuracy with the same data
- multi-layer perceptron (MLP) architecture: criteria for choosing number of hidden layers and size of the hidden layer? [closed]
- What is the role of TimeDistributed layer in Keras?
- How to interpret loss and accuracy for a machine learning model [closed]
- Can Keras deal with input images with different size?
- TimeDistributed(Dense) vs Dense in Keras – Same number of parameters
- Keras: model.evaluate vs model.predict accuracy difference in multi-class NLP task
- Caffe: What can I do if only a small batch fits into memory?
- How to create caffe.deploy from train.prototxt
- How to reduce a fully-connected (`”InnerProduct”`) layer using truncated SVD
- keras: how to save the training history attribute of the history object
- What is `lr_policy` in Caffe?
- Using pre-trained word2vec with LSTM for word generation
- Choosing number of Steps per Epoch
- Is this possible to predict the lottery numbers (not the most accurate)? [closed]
- Why must a nonlinear activation function be used in a backpropagation neural network? [closed]
- How to fix “ResourceExhaustedError: OOM when allocating tensor”