This question was raised at the Keras github repository in Issue #4446: Quick Question: can a model be fit for multiple times? It was closed by François Chollet with the following statement:

Yes, successive calls to

`fit`

will incrementally train the model.

So, yes, you can call fit multiple times.

### Related Contents:

- Why binary_crossentropy and categorical_crossentropy give different performances for the same problem?
- What is the role of “Flatten” in Keras?
- Many to one and many to many LSTM examples in Keras
- What’s the difference between a bidirectional LSTM and an LSTM?
- How to tell Keras stop training based on loss value?
- Instance Normalisation vs Batch normalisation
- What is the difference between loss function and metric in Keras? [duplicate]
- Keras model.summary() result – Understanding the # of Parameters
- Keras Text Preprocessing – Saving Tokenizer object to file for scoring
- What is the difference between Keras model.evaluate() and model.predict()?
- What is the number of filter in CNN?
- Difference between Keras model.save() and model.save_weights()?
- What is the difference between an Embedding Layer and a Dense Layer?
- What is the role of the bias in neural networks? [closed]
- Epoch vs Iteration when training neural networks [closed]
- What are advantages of Artificial Neural Networks over Support Vector Machines? [closed]
- How to interpret loss and accuracy for a machine learning model [closed]
- Intuitive understanding of 1D, 2D, and 3D convolutions in convolutional neural networks [closed]
- Why do we have to normalize the input for an artificial neural network? [closed]
- How to train an artificial neural network to play Diablo 2 using visual input?
- pytorch – connection between loss.backward() and optimizer.step()
- Deep-Learning Nan loss reasons
- Why should weights of Neural Networks be initialized to random numbers? [closed]
- NaN loss when training regression network
- Common causes of nans during training of neural networks
- multi-layer perceptron (MLP) architecture: criteria for choosing number of hidden layers and size of the hidden layer? [closed]
- How to concatenate two layers in keras?
- What is the role of TimeDistributed layer in Keras?
- keras: how to save the training history attribute of the history object
- Keras: Difference between Kernel and Activity regularizers
- Batch Normalization in Convolutional Neural Network
- How to update the bias in neural network backpropagation?
- Higher validation accuracy, than training accurracy using Tensorflow and Keras
- Calculate the output size in convolution layer [closed]
- TensorFlow – regularization with L2 loss, how to apply to all weights, not just last one?
- What is a multi-headed model? And what exactly is a ‘head’ in a model?
- Convert Keras model to C++ [closed]
- Open Source Neural Network Library [closed]
- How to log Keras loss output to a file
- Keras accuracy does not change
- What is a projection layer in the context of neural networks?
- What is the correct way to change image channel ordering between channels first and channels last?
- How does one debug NaN values in TensorFlow?
- CNN – Image Resizing VS Padding (keeping aspect ratio or not?)
- In which cases is the cross-entropy preferred over the mean squared error? [closed]
- Meaning of an Epoch in Neural Networks Training
- What does the standard Keras model output mean? What is epoch and loss in Keras?
- Can neural networks approximate any function given enough hidden neurons?
- What is the difference between UpSampling2D and Conv2DTranspose functions in keras?
- What is the use of train_on_batch() in keras?