Xc@@sddlmZddlmZddlZddlZddlZddlZddlZddlZ ddl m Z ddl m Z ddl mZddlmZdd lmZdd lmZdd lmZdd lmZddlmZdd lmZddlmZyddlZWnek rAdZnXe!dZ"ddZ#ddZ$ddZ%ddZ&defdYZ'dS(i(tabsolute_import(tprint_functionNi(tbackend(t optimizers(tlayers(task_to_proceed_with_overwrite(tModel(ttopology(tLayer(tInput(tmodels(t interfacescC@sstdkrtdnd}ddlm}| rftjj|rft|}|sfdSntj |d}t |j d|j dR?R@RJt get_weightsROR0R(RRR(R((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRs  cC@stj|r\tj|}x;|D]0}t|j}|j|| ||}q%Wn|jdkrx|jn|jj|dS(sSets the weights of the model. # Arguments weights: Should be a list of Numpy arrays with shapes and types matching the output of `model.get_weights()`. N( R>R?R@RR(RqROR0R(RR(RRtnb_param((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRqs   cC@stdkrtdntj|dd}d|jkrXd|krX|d}ntj|rytj|}n |j}|rt j ||nt j ||t |dr|j ndS(Ns`load_weights` requires h5py.ReRft layer_namesR RN(R/R0R1R6R9R>R?R@RRt$load_weights_from_hdf5_group_by_nameRlRRN(RRPtby_nameRSR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyt load_weightss   cC@stdkrtdn| rMtjj|rMt|}|sMdSntj|rntj |}n |j }tj |d}t j |||j|jdS(Ns`save_weights` requires h5py.R(R/R0R1R3R4R5RR>R?R@RR6RRARMRN(RRPRQRRRRS((s+/tmp/pip-build-isqEY4/keras/keras/models.pyt save_weightss    cK@s|j|jj||d|d|||jj|_|jj|_|jj|_|jj|_|jj|_|jj|_|jj |_ dS(sConfigures the learning process. # Arguments optimizer: str (name of optimizer) or optimizer object. See [optimizers](/optimizers). loss: str (name of objective function) or objective function. See [objectives](/objectives). metrics: list of metrics to be evaluated by the model during training and testing. Typically you will use `metrics=['accuracy']`. See [metrics](/metrics). sample_weight_mode: if you need to do timestep-wise sample weighting (2D weights), set this to "temporal". "None" defaults to sample-wise weights (1D). **kwargs: for Theano backend, these are passed into K.function. Ignored for Tensorflow backend. # Example ```python model = Sequential() model.add(Dense(32, input_shape=(500,))) model.add(Dense(10, activation='softmax')) model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy']) ``` R$R%N( RRORnR!R#R&R$tmetrics_tensorst metrics_namesR%(RR!R#R$R%tkwargs((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRns i i igic K@sd| kr+tjd| jd}n| rJtdt| n|jdkrhtdn|jj||d|d|d|d|d |d |d | d | d | d|  S(s Trains the model for a fixed number of epochs. # Arguments x: input data, as a Numpy array or list of Numpy arrays (if the model has multiple inputs). y: labels, as a Numpy array. batch_size: integer. Number of samples per gradient update. epochs: integer, the number of epochs to train the model. verbose: 0 for no logging to stdout, 1 for progress bar logging, 2 for one log line per epoch. callbacks: list of `keras.callbacks.Callback` instances. List of callbacks to apply during training. See [callbacks](/callbacks). validation_split: float (0. < x < 1). Fraction of the data to use as held-out validation data. validation_data: tuple (x_val, y_val) or tuple (x_val, y_val, val_sample_weights) to be used as held-out validation data. Will override validation_split. shuffle: boolean or str (for 'batch'). Whether to shuffle the samples at each epoch. 'batch' is a special option for dealing with the limitations of HDF5 data; it shuffles in batch-sized chunks. class_weight: dictionary mapping classes to a weight value, used for scaling the loss function (during training only). sample_weight: Numpy array of weights for the training samples, used for scaling the loss function (during training only). You can either pass a flat (1D) Numpy array with the same length as the input samples (1:1 mapping between weights and samples), or in the case of temporal data, you can pass a 2D array with shape (samples, sequence_length), to apply a different weight to every timestep of every sample. In this case you should make sure to specify sample_weight_mode="temporal" in compile(). initial_epoch: epoch at which to start training (useful for resuming a previous training run) # Returns A `History` object. Its `History.history` attribute is a record of training loss values and metrics values at successive epochs, as well as validation loss values and validation metrics values (if applicable). # Raises RuntimeError: if the model was never compiled. tnb_epochs;The `nb_epoch` argument in `fit` has been renamed `epochs`.s Unrecognized keyword arguments: s1The model needs to be compiled before being used.t batch_sizetepochstverboset callbackstvalidation_splittvalidation_datatshufflet class_weightt sample_weightt initial_epochN( RDRERRR7ROR0t RuntimeErrortfit(RRtyRRRRRRRRRRR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRs$2  c C@sC|jdkrtdn|jj||d|d|d|S(sComputes the loss on some input data, batch by batch. # Arguments x: input data, as a Numpy array or list of Numpy arrays (if the model has multiple inputs). y: labels, as a Numpy array. batch_size: integer. Number of samples per gradient update. verbose: verbosity mode, 0 or 1. sample_weight: sample weights, as a Numpy array. # Returns Scalar test loss (if the model has no metrics) or list of scalars (if the model computes other metrics). The attribute `model.metrics_names` will give you the display labels for the scalar outputs. # Raises RuntimeError: if the model was never compiled. s1The model needs to be compiled before being used.RRRN(ROR0Rtevaluate(RRRRRR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyROs cC@s8|jdkr|jn|jj|d|d|S(sFGenerates output predictions for the input samples. The input samples are processed batch by batch. # Arguments x: the input data, as a Numpy array. batch_size: integer. verbose: verbosity mode, 0 or 1. # Returns A Numpy array of predictions. RRN(ROR0Rtpredict(RRRR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRls  cC@s,|jdkr|jn|jj|S(sReturns predictions for a single batch of samples. # Arguments x: input data, as a Numpy array or list of Numpy arrays (if the model has multiple inputs). # Returns A Numpy array of predictions. N(ROR0Rtpredict_on_batch(RR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyR}s  cC@s=|jdkrtdn|jj||d|d|S(sSingle gradient update over one batch of samples. # Arguments x: input data, as a Numpy array or list of Numpy arrays (if the model has multiple inputs). y: labels, as a Numpy array. class_weight: dictionary mapping classes to a weight value, used for scaling the loss function (during training only). sample_weight: sample weights, as a Numpy array. # Returns Scalar training loss (if the model has no metrics) or list of scalars (if the model computes other metrics). The attribute `model.metrics_names` will give you the display labels for the scalar outputs. # Raises RuntimeError: if the model was never compiled. s1The model needs to be compiled before being used.RRN(ROR0Rttrain_on_batch(RRRRR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRs cC@s7|jdkrtdn|jj||d|S(sEvaluates the model over a single batch of samples. # Arguments x: input data, as a Numpy array or list of Numpy arrays (if the model has multiple inputs). y: labels, as a Numpy array. sample_weight: sample weights, as a Numpy array. # Returns Scalar test loss (if the model has no metrics) or list of scalars (if the model computes other metrics). The attribute `model.metrics_names` will give you the display labels for the scalar outputs. # Raises RuntimeError: if the model was never compiled. s1The model needs to be compiled before being used.RN(ROR0Rt test_on_batch(RRRR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRscC@sM|j|||}|jdks9|jdkrItjdn|S(sGenerates class probability predictions for the input samples. The input samples are processed batch by batch. # Arguments x: input data, as a Numpy array or list of Numpy arrays (if the model has multiple inputs). batch_size: integer. verbose: verbosity mode, 0 or 1. # Returns A Numpy array of probability predictions. gg?sNetwork returning invalid probability values. The last layer might not normalize predictions into probabilities (like softmax or sigmoid would).(RtmintmaxRDRE(RRRRtpreds((s+/tmp/pip-build-isqEY4/keras/keras/models.pyt predict_probas$cC@sU|j|d|d|}|jddkr>|jddS|dkjdSdS( sGenerate class predictions for the input samples. The input samples are processed batch by batch. # Arguments x: input data, as a Numpy array or list of Numpy arrays (if the model has multiple inputs). batch_size: integer. verbose: verbosity mode, 0 or 1. # Returns A numpy array of class predictions. RRiitaxisg?tint32N(RRLtargmaxtastype(RRRRtproba((s+/tmp/pip-build-isqEY4/keras/keras/models.pytpredict_classessc C@sj|jd krtdn|jj|||d|d|d|d|d|d| d| d | d |  S( s Fits the model on data generated batch-by-batch by a Python generator. The generator is run in parallel to the model, for efficiency. For instance, this allows you to do real-time data augmentation on images on CPU in parallel to training your model on GPU. # Arguments generator: A generator. The output of the generator must be either - a tuple (inputs, targets) - a tuple (inputs, targets, sample_weights). All arrays should contain the same number of samples. The generator is expected to loop over its data indefinitely. An epoch finishes when `samples_per_epoch` samples have been seen by the model. steps_per_epoch: Total number of steps (batches of samples) to yield from `generator` before declaring one epoch finished and starting the next epoch. It should typically be equal to the number of unique samples of your dataset divided by the batch size. epochs: Integer, total number of iterations on the data. verbose: Verbosity mode, 0, 1, or 2. callbacks: List of callbacks to be called during training. validation_data: This can be either - A generator for the validation data - A tuple (inputs, targets) - A tuple (inputs, targets, sample_weights). validation_steps: Only relevant if `validation_data` is a generator. Number of samples to use from validation generator at the end of every epoch. class_weight: Dictionary mapping class indices to a weight for the class. max_q_size: Maximum size for the generator queue workers: Maximum number of processes to spin up pickle_safe: Ff True, use process based threading. Note that because this implementation relies on multiprocessing, you should not pass non picklable arguments to the generator as they can't be passed easily to children processes. initial_epoch: Epoch at which to start training (useful for resuming a previous training run) # Returns A `History` object. # Raises RuntimeError: if the model was never compiled. # Example ```python def generate_arrays_from_file(path): while 1: f = open(path) for line in f: # create Numpy arrays of input data # and labels, from each line in the file x, y = process_line(line) yield (x, y) f.close() model.fit_generator(generate_arrays_from_file('/my_file.txt'), samples_per_epoch=10000, epochs=10) ``` s1The model needs to be compiled before being used.RRRtvalidation_stepsRt max_q_sizetworkerst pickle_safeRN(ROR0Rt fit_generator( Rt generatortsteps_per_epochRRRRRRRRRR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRsQ c C@sC|jdkrtdn|jj||d|d|d|S(sEvaluates the model on a data generator. The generator should return the same kind of data as accepted by `test_on_batch`. # Arguments generator: Generator yielding tuples (inputs, targets) or (inputs, targets, sample_weights) steps: Total number of steps (batches of samples) to yield from `generator` before stopping. max_q_size: maximum size for the generator queue workers: maximum number of processes to spin up pickle_safe: if True, use process based threading. Note that because this implementation relies on multiprocessing, you should not pass non picklable arguments to the generator as they can't be passed easily to children processes. # Returns Scalar test loss (if the model has no metrics) or list of scalars (if the model computes other metrics). The attribute `model.metrics_names` will give you the display labels for the scalar outputs. # Raises RuntimeError: if the model was never compiled. s1The model needs to be compiled before being used.RRRN(ROR0Rtevaluate_generator(RRtstepsRRR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRKs c C@sG|jdkr|jn|jj||d|d|d|d|S(sGenerates predictions for the input samples from a data generator. The generator should return the same kind of data as accepted by `predict_on_batch`. # Arguments generator: generator yielding batches of input samples. steps: Total number of steps (batches of samples) to yield from `generator` before stopping. max_q_size: maximum size for the generator queue workers: maximum number of processes to spin up pickle_safe: if True, use process based threading. Note that because this implementation relies on multiprocessing, you should not pass non picklable arguments to the generator as they can't be passed easily to children processes. verbose: verbosity mode, 0 or 1. # Returns A Numpy array of predictions. RRRRN(ROR0Rtpredict_generator(RRRRRRR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRss cC@sqt|jdtjr#|jSg}x8|jD]-}|ji|jjd6|jd6q3Wt j |S(NiR R( RBRRRtlegacy_get_configRJRRR tcopytdeepcopy(RRR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyR s cC@skd|dks$|dddkr1|j|S|}x*|D]"}tj|}|j|qAW|S(NR iR(tlegacy_from_configRvRmR(tclsRROtconfR((s+/tmp/pip-build-isqEY4/keras/keras/models.pyt from_configs$   cC@sEg}t|jdtjrt|jdds;tg}xE|jdjD]3}i|jjd6|jd6}|j |qRW|jdj}||d<|j idd6|d6n5|j i|jdjjd6|jdjd6x<|jdD]-}|j i|jjd6|jd6qWt j |S(sRetrieves the model configuration as a Python list. # Returns A list of dicts (each dict is a layer config). iRR RRi( RBRRRRtAssertionErrorRRR RJRR(RRRRt layer_configt merge_config((s+/tmp/pip-build-isqEY4/keras/keras/models.pyRs" c@ssind}|}fd}|d}||}|ddkrg}|d}x3|jdD]"} tj| } |j| qvW||dR R/R1R0RR\RuRkRzR|Ro(((s+/tmp/pip-build-isqEY4/keras/keras/models.pyts8         l