python类savez()的实例源码

def writeToFile(self, filename, saveOpts=False):
        '''
        write the distortion coeffs to file
        saveOpts --> Whether so save calibration options (and not just results)
        '''
        try:
            if not filename.endswith('.%s' % self.ftype):
                filename += '.%s' % self.ftype
            s = {'coeffs': self.coeffs}
            if saveOpts:
                s['opts'] = self.opts
#             else:
#                 s['opts':{}]
            np.savez(filename, **s)
            return filename
        except AttributeError:
            raise Exception(
                'need to calibrate camera before calibration can be saved to file')

def _save_model(self, bleu_score):
        if self._is_valid_to_save(bleu_score):
            model = ModelInfo(bleu_score, self.config['saveto'])

            # Manage n-best model list first
            if len(self.best_models) >= self.track_n_models:
                old_model = self.best_models[0]
                if old_model.path and os.path.isfile(old_model.path):
                    logger.info("Deleting old model %s" % old_model.path)
                    os.remove(old_model.path)
                self.best_models.remove(old_model)

            self.best_models.append(model)
            self.best_models.sort(key=operator.attrgetter('bleu_score'))

            # Save the model here
            s = signal.signal(signal.SIGINT, signal.SIG_IGN)
            logger.info("Saving new model {}".format(model.path))
            numpy.savez(
                model.path, **self.main_loop.model.get_parameter_dict())
            numpy.savez(
                os.path.join(self.config['saveto'], 'val_bleu_scores.npz'),
                bleu_scores=self.val_bleu_curve)
            signal.signal(signal.SIGINT, s)

def save_training(self, filename):
        """
        Save traning set and labels of current model
        :param filename: filename of new data.npz, it will be saved in 'train/'
        """
        np.savez(os.path.join(TRAIN_DATA, filename),
                 train_set=self.train_set,
                 train_labels=self.train_labels)

def end(self, _session):
    np.savez(self.params["file"], **self._beam_accum)

def end(self, _session):
    scores_path = os.path.join(self.params["output_dir"],
                               "attention_scores.npz")
    np.savez(scores_path, *self._attention_scores_accum)
    tf.logging.info("Wrote %s", scores_path)

def save_path(self, path):
        self._model.save(path)
        self._feature_scaler.save_path(path.with_suffix('.feature_scaler'))

        with open(path.with_suffix('.pessimism'), 'wb') as f:
            np.savez(
                f,
                aim_pessimism_factor=self._aim_pessimism_factor,
                accuracy_pessimism_factor=self._accuracy_pessimism_factor,
            )

def save_path(self, path):
        with open(path, 'wb') as f:
            np.savez(
                f,
                ndim=len(self._axes) + 1,
                mean=self.mean,
                std=self.std,
            )

def dataset_to_file(dataset, ngrams, filename='dataset'):
    """Save a dataset to a file.

    Args:
        dataset (:class:`np.ndarray`): the dataset to save (built with :func:`dataset_tools.build_dataset`)
        ngrams (list of strings): the ngrams used to compute the features
        filename (string): the filename without extension (will be .npz)
    """
    num_samples, num_entries, num_features = dataset.shape
    # We rehaspe the ndarray from 3D to 2D in order to write it into a text file
    # Each line of the file will correspond to one cited paper
    # Therefore, on each there will be the `num_entries` sets of features
    dataset_sp = sparse.csr_matrix(dataset.reshape(num_samples*num_entries, num_features))
    np.savez(filename, num_entries=np.array([num_entries]), data=dataset_sp.data, indices=dataset_sp.indices,
             indptr=dataset_sp.indptr, shape=dataset_sp.shape, ngrams=ngrams)

def save_sparse_csr(filename,array):
    np.savez(filename,data = array.data ,indices=array.indices,
             indptr =array.indptr, shape=array.shape )

def secure_numpy_save(params_dict, path):
    """Try saving into a temporary file and then move."""
    try:
        dirname = os.path.dirname(path)
        with tempfile.NamedTemporaryFile(delete=False, dir=dirname) as temp:
            numpy.savez(temp, **params_dict)
        shutil.move(temp.name, path)
    except Exception as e:
        # if "temp" in locals():
        #    os.remove(temp.name)
        logger.error(" Error {0}".format(str(e)))

def _save_params(self, model, params):

        # Rename accordingly for blocks compatibility
        params_to_save = dict(
            (k.replace('/', '-'), v) for k, v in params.items())

        numpy.savez(model.path, **params_to_save)

def _save_bleu_scores(self):
        numpy.savez(
            os.path.join(
                self.saveto,
                'val_bleu_scores{}_{}.npz'.format(self.enc_id, self.dec_id)),
            bleu_scores=self.val_bleu_curve)

def encode_npz(subvol):
    """
    This file format is unrelated to np.savez
    We are just saving as .npy and the compressing
    using zlib. 
    The .npy format contains metadata indicating
    shape and dtype, instead of np.tobytes which doesn't
    contain any metadata.
    """
    fileobj = io.BytesIO()
    if len(subvol.shape) == 3:
        subvol = np.expand_dims(subvol, 0)
    np.save(fileobj, subvol)
    cdz = zlib.compress(fileobj.getvalue())
    return cdz

def save_np(reader, output_filename):
    ts, data = reader.read_stack()
    np.savez(output_filename, timestamps=ts, data=data)

def write_test_file(self, variable='v', check=False):
        data, metadata = self.build_test_data(variable)
        metadata_array = np.array(sorted(metadata.items()))
        np.savez(self.test_file, data=data, metadata=metadata_array)
        if check:
            data1, metadata1 = read_test_file(self.test_file)
            assert metadata == metadata1, "%s != %s" % (metadata, metadata1)
            assert data.shape == data1.shape == (505, 2), \
                "%s, %s, (505, 2)" % (data.shape, data1.shape)
            assert (data == data1).all()
            assert metadata["n"] == 505

def _write_file_contents(self, data, metadata):
        # we explicitly set the dtype to ensure roundtrips preserve file contents exactly
        max_metadata_length = max(chain([len(k) for k in metadata.keys()],
                                        [len(str(v)) for v in metadata.values()]))
        if PY2:
            dtype = "S%d" % max_metadata_length
        else:
            dtype = "U%d" % max_metadata_length
        metadata_array = numpy.array(sorted(metadata.items()), dtype)
        numpy.savez(self.filename, data=data, metadata=metadata_array)

def write_test_file(self, variable='v', check=False):
        data, metadata = self.build_test_data(variable)
        metadata_array = np.array(sorted(metadata.items()))
        np.savez(self.test_file, data=data, metadata=metadata_array)
        if check:
            data1, metadata1 = read_test_file(self.test_file)
            assert metadata == metadata1, "%s != %s" % (metadata, metadata1)
            assert data.shape == data1.shape == (505, 2), \
                "%s, %s, (505, 2)" % (data.shape, data1.shape)
            assert (data == data1).all()
            assert metadata["n"] == 505

def _write_file_contents(self, data, metadata):
        # we explicitly set the dtype to ensure roundtrips preserve file contents exactly
        max_metadata_length = max(chain([len(k) for k in metadata.keys()],
                                        [len(str(v)) for v in metadata.values()]))
        if PY2:
            dtype = "S%d" % max_metadata_length
        else:
            dtype = "U%d" % max_metadata_length
        metadata_array = numpy.array(sorted(metadata.items()), dtype)
        numpy.savez(self.filename, data=data, metadata=metadata_array)

def save_model(self, file_name, directory):
        """Saves a checkpoint of the model and a .npz file with stored rewards.

        Args:
        file_name: String name to use for the checkpoint and rewards files.
        Defaults to self.model_name if None is provided.
        """
        if self.verbose: print "Saving model..."

        save_dir = directory + file_name
        os.mkdir(save_dir)
        directory = save_dir + '/'

        save_loc = os.path.join(directory, file_name + '.ckpt')
        training_epochs = len(self.training_val_results) * self.accuracy_logged_every_n
        self.saver.save(self.session, save_loc, global_step=training_epochs)


        npz_name = os.path.join(directory, file_name + '-' + str(training_epochs))

        if not self.print_per_task:
            np.savez(npz_name,
                    training_val_results=self.training_val_results,
                    l2_beta=self.l2_beta,
                    dropout=self.dropout,
                    hidden_sizes_shared=self.hidden_sizes_shared,
                    hidden_size_task=self.hidden_size_task)

def save_model(self, file_name, directory):
        """Saves a checkpoint of the model and a .npz file with stored rewards.

        Args:
        file_name: String name to use for the checkpoint and rewards files.
        Defaults to self.model_name if None is provided.
        """
        if self.verbose: print "Saving model..."

        save_dir = directory + file_name
        os.mkdir(save_dir)
        directory = save_dir + '/'

        save_loc = os.path.join(directory, file_name + '.ckpt')
        training_epochs = len(self.training_val_results) * self.accuracy_logged_every_n
        self.saver.save(self.session, save_loc, global_step=training_epochs)


        npz_name = os.path.join(directory, file_name + '-' + str(training_epochs))

        if not self.print_per_task:
            np.savez(npz_name,
                    training_val_results=self.training_val_results,
                    l2_beta=self.l2_beta,
                    dropout=self.dropout,
                    hidden_sizes_shared=self.hidden_sizes_shared,
                    hidden_size_task=self.hidden_size_task)
        else:
            np.savez(npz_name,
                    training_val_results=self.training_val_results,
                    training_val_results_per_task=self.training_val_results_per_task,
                    l2_beta=self.l2_beta,
                    dropout=self.dropout,
                    hidden_sizes_shared=self.hidden_sizes_shared,
                    hidden_size_task=self.hidden_size_task)