MCPcopy Create free account
hub / github.com/DeepRec-AI/DeepRec / set_callback_parameters

Function set_callback_parameters

tensorflow/python/keras/callbacks.py:122–166  ·  view source on GitHub ↗

Sets callback parameters. Arguments: callback_list: CallbackList instance. model: Model being trained. do_validation: Whether or not validation loop will be run. batch_size: Number of samples per batch. epochs: Number of epoch to train. steps_per_epoch: Number

(callback_list,
                            model,
                            do_validation=False,
                            batch_size=None,
                            epochs=None,
                            steps_per_epoch=None,
                            samples=None,
                            verbose=1,
                            mode=ModeKeys.TRAIN)

Source from the content-addressed store, hash-verified

120
121
122def set_callback_parameters(callback_list,
123 model,
124 do_validation=False,
125 batch_size=None,
126 epochs=None,
127 steps_per_epoch=None,
128 samples=None,
129 verbose=1,
130 mode=ModeKeys.TRAIN):
131 """Sets callback parameters.
132
133 Arguments:
134 callback_list: CallbackList instance.
135 model: Model being trained.
136 do_validation: Whether or not validation loop will be run.
137 batch_size: Number of samples per batch.
138 epochs: Number of epoch to train.
139 steps_per_epoch: Number of batches to run per training epoch.
140 samples: Number of training samples.
141 verbose: int, 0 or 1. Keras logging verbosity to pass to ProgbarLogger.
142 mode: String. One of ModeKeys.TRAIN, ModeKeys.TEST, or ModeKeys.PREDICT.
143 Which loop mode to configure callbacks for.
144 """
145 for cbk in callback_list:
146 if isinstance(cbk, (BaseLogger, ProgbarLogger)):
147 cbk.stateful_metrics = model.metrics_names[1:] # Exclude `loss`
148
149 # Set callback parameters
150 callback_metrics = []
151 # When we have deferred build scenario with iterator input, we will compile
152 # when we standardize first batch of data.
153 if mode != ModeKeys.PREDICT and hasattr(model, 'metrics_names'):
154 callback_metrics = copy.copy(model.metrics_names)
155 if do_validation:
156 callback_metrics += ['val_' + n for n in model.metrics_names]
157 callback_params = {
158 'batch_size': batch_size,
159 'epochs': epochs,
160 'steps': steps_per_epoch,
161 'samples': samples,
162 'verbose': verbose,
163 'do_validation': do_validation,
164 'metrics': callback_metrics,
165 }
166 callback_list.set_params(callback_params)
167
168
169def _is_generator_like(data):

Callers 1

configure_callbacksFunction · 0.85

Calls 2

copyMethod · 0.45
set_paramsMethod · 0.45

Tested by

no test coverage detected