How to load a model from an HDF5 file in Keras

Question

How to load a model from an HDF5 file in Keras   What I tried   model   Sequential    model add Dense 64  input dim 14  init  uniform    model add LeakyReLU alpha 0 3   model add BatchNormalization epsilon 1e-06  mode 0  momentum 0 9  weights None   model add Dropout 0 5    model add Dense 64  init  uniform    model add LeakyReLU alpha 0 3   model add BatchNormalization epsilon 1e-06  mode 0  momentum 0 9  weights None   model add Dropout 0 5    model add Dense 2  init  uniform    model add Activation  softmax      sgd   SGD lr 0 1  decay 1e-6  momentum 0 9  nesterov True  model compile loss  binary crossentropy   optimizer sgd   checkpointer   ModelCheckpoint filepath   weights hdf5   verbose 1  save best only True  model fit X train  y train  nb epoch 20  batch size 16  show accuracy True  validation split 0 2  verbose   2  callbacks  checkpointer     The above code successfully saves the best model to a file named weights hdf5  What I want to do is then load that model  The below code shows how I tried to do so   model2   Sequential   model2 load weights   Users Desktop SquareSpace weights hdf5     This is the error I get   IndexError                                Traceback  most recent call last   lt ipython-input-101-ec968f9e95c5 gt  in  lt module gt          1 model2   Sequential   ---- gt  2 model2 load weights   Users Desktop SquareSpace weights hdf5     Applications anaconda lib python2 7 site-packages keras models pyc in load weights self  filepath      582             g   f  layer     format k       583             weights    g  param     format p   for p in range g attrs  nb params     -- gt  584             self layers k  set weights weights      585         f close       586   IndexError  list index out of range

User · Answer

According to official documentation https://keras.io/getting-started/faq/#how-can-i-install-hdf5-or-h5py-to-save-my-models-in-keras

you can do :

first test if you have h5py installed by running the

import h5py

if you dont have errors while importing h5py you are good to save:

from keras.models import load_model

model.save('my_model.h5')  # creates a HDF5 file 'my_model.h5'
del model  # deletes the existing model

# returns a compiled model
# identical to the previous one
model = load_model('my_model.h5')

If you need to install h5py http://docs.h5py.org/en/latest/build.html

User · Answer

load weights only sets the weights of your network  You still need to define its architecture before calling load weights   def create model       model   Sequential      model add Dense 64  input dim 14  init  uniform       model add LeakyReLU alpha 0 3      model add BatchNormalization epsilon 1e-06  mode 0  momentum 0 9  weights None      model add Dropout 0 5       model add Dense 64  init  uniform       model add LeakyReLU alpha 0 3      model add BatchNormalization epsilon 1e-06  mode 0  momentum 0 9  weights None      model add Dropout 0 5      model add Dense 2  init  uniform       model add Activation  softmax       return model  def train       model   create model      sgd   SGD lr 0 1  decay 1e-6  momentum 0 9  nesterov True     model compile loss  binary crossentropy   optimizer sgd      checkpointer   ModelCheckpoint filepath   tmp weights hdf5   verbose 1  save best only True     model fit X train  y train  nb epoch 20  batch size 16  show accuracy True  validation split 0 2  verbose 2  callbacks  checkpointer    def load trained model weights path      model   create model      model load weights weights path

User · Answer

I done in this way  from keras models import Sequential from keras contrib losses import import crf loss from keras contrib metrics import crf viterbi accuracy    To save model model save  my model 01 hdf5      To load the model custom objects   CRF   CRF  crf loss   crf loss  crf viterbi accuracy  crf viterbi accuracy     To load a persisted model that uses the CRF layer  model1   load model   home abc my model 01 hdf5   custom objects   custom objects

User · Answer

See the following sample code on how to Build a basic Keras Neural Net Model  save Model  JSON   amp  Weights  HDF5  and load them     create model model   Sequential   model add Dense X shape 1   input dim X shape 1   activation  relu     Input Layer model add Dense X shape 1   activation  relu     Hidden Layer model add Dense output dim  activation  softmax     Output Layer    Compile  amp  Fit model model compile loss  binary crossentropy   optimizer  adam   metrics   accuracy    model fit X Y nb epoch 5 batch size 100 verbose 1         serialize model to JSON model json   model to json   with open  Data model json    w   as json file      json file write simplejson dumps simplejson loads model json   indent 4      serialize weights to HDF5 model save weights  Data model h5   print  Saved model to disk      load json and create model json file   open  Data model json    r   loaded model json   json file read   json file close   loaded model   model from json loaded model json     load weights into new model loaded model load weights  Data model h5   print  Loaded model from disk      evaluate loaded model on test data    Define X test  amp  Y test data first loaded model compile loss  binary crossentropy   optimizer  adam   metrics   accuracy    score   loaded model evaluate X test  Y test  verbose 0  print    s    2f       loaded model metrics names 1   score 1  100

User · Answer

If you stored the complete model  not only the weights  in the HDF5 file  then it is as simple as  from keras models import load model model   load model  model h5

[python] How to load a model from an HDF5 file in Keras?

Examples related to python

Examples related to machine-learning

Examples related to keras

Examples related to data-science