Python numpy.float32() Examples

The following are 30 code examples for showing how to use numpy.float32(). These examples are extracted from open source projects. You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example.

You may check out the related API usage on the sidebar.

You may also want to check out all available functions/classes of the module numpy , or try the search function .

Example 1
Project: Black-Box-Audio   Author: rtaori   File: run_audio_attack.py    License: MIT License 6 votes vote down vote up
def __init__(self, input_wave_file, output_wave_file, target_phrase):
        self.pop_size = 100
        self.elite_size = 10
        self.mutation_p = 0.005
        self.noise_stdev = 40
        self.noise_threshold = 1
        self.mu = 0.9
        self.alpha = 0.001
        self.max_iters = 3000
        self.num_points_estimate = 100
        self.delta_for_gradient = 100
        self.delta_for_perturbation = 1e3
        self.input_audio = load_wav(input_wave_file).astype(np.float32)
        self.pop = np.expand_dims(self.input_audio, axis=0)
        self.pop = np.tile(self.pop, (self.pop_size, 1))
        self.output_wave_file = output_wave_file
        self.target_phrase = target_phrase
        self.funcs = self.setup_graph(self.pop, np.array([toks.index(x) for x in target_phrase])) 
Example 2
Project: cat-bbs   Author: aleju   File: bbs.py    License: MIT License 6 votes vote down vote up
def draw_on_image(self, img, color=[0, 255, 0], alpha=1.0, copy=True, from_img=None):
        if copy:
            img = np.copy(img)

        orig_dtype = img.dtype
        if alpha != 1.0 and img.dtype != np.float32:
            img = img.astype(np.float32, copy=False)

        for rect in self:
            if from_img is not None:
                rect.resize(from_img, img).draw_on_image(img, color=color, alpha=alpha, copy=False)
            else:
                rect.draw_on_image(img, color=color, alpha=alpha, copy=False)

        if orig_dtype != img.dtype:
            img = img.astype(orig_dtype, copy=False)

        return img 
Example 3
Project: cat-bbs   Author: aleju   File: bbs.py    License: MIT License 6 votes vote down vote up
def draw_on_image(self, img, color=[0, 255, 0], alpha=1.0, copy=True):
        if copy:
            img = np.copy(img)

        orig_dtype = img.dtype
        if alpha != 1.0 and img.dtype != np.float32:
            img = img.astype(np.float32, copy=False)

        for rectimg in self:
            if rectimg is not None:
                rectimg.draw_on_image(img, color=color, alpha=alpha, copy=False)

        if orig_dtype != img.dtype:
            img = img.astype(orig_dtype, copy=False)

        return img 
Example 4
Project: cat-bbs   Author: aleju   File: common.py    License: MIT License 6 votes vote down vote up
def draw_heatmap(img, heatmap, alpha=0.5):
    """Draw a heatmap overlay over an image."""
    assert len(heatmap.shape) == 2 or \
        (len(heatmap.shape) == 3 and heatmap.shape[2] == 1)
    assert img.dtype in [np.uint8, np.int32, np.int64]
    assert heatmap.dtype in [np.float32, np.float64]

    if img.shape[0:2] != heatmap.shape[0:2]:
        heatmap_rs = np.clip(heatmap * 255, 0, 255).astype(np.uint8)
        heatmap_rs = ia.imresize_single_image(
            heatmap_rs[..., np.newaxis],
            img.shape[0:2],
            interpolation="nearest"
        )
        heatmap = np.squeeze(heatmap_rs) / 255.0

    cmap = plt.get_cmap('jet')
    heatmap_cmapped = cmap(heatmap)
    heatmap_cmapped = np.delete(heatmap_cmapped, 3, 2)
    heatmap_cmapped = heatmap_cmapped * 255
    mix = (1-alpha) * img + alpha * heatmap_cmapped
    mix = np.clip(mix, 0, 255).astype(np.uint8)
    return mix 
Example 5
Project: aospy   Author: spencerahill   File: data_loader.py    License: Apache License 2.0 6 votes vote down vote up
def _maybe_cast_to_float64(da):
    """Cast DataArrays to np.float64 if they are of type np.float32.

    Parameters
    ----------
    da : xr.DataArray
        Input DataArray

    Returns
    -------
    DataArray

    """
    if da.dtype == np.float32:
        logging.warning('Datapoints were stored using the np.float32 datatype.'
                        'For accurate reduction operations using bottleneck, '
                        'datapoints are being cast to the np.float64 datatype.'
                        ' For more information see: https://github.com/pydata/'
                        'xarray/issues/1346')
        return da.astype(np.float64)
    else:
        return da 
Example 6
Project: Att-ChemdNER   Author: lingluodlut   File: utils.py    License: Apache License 2.0 6 votes vote down vote up
def set_values(name, param, pretrained):
#{{{
    """
    Initialize a network parameter with pretrained values.
    We check that sizes are compatible.
    """
    param_value = param.get_value()
    if pretrained.size != param_value.size:
        raise Exception(
            "Size mismatch for parameter %s. Expected %i, found %i."
            % (name, param_value.size, pretrained.size)
        )
    param.set_value(np.reshape(
        pretrained, param_value.shape
    ).astype(np.float32))
#}}} 
Example 7
Project: Att-ChemdNER   Author: lingluodlut   File: optimization.py    License: Apache License 2.0 6 votes vote down vote up
def sgd(self, cost, params,constraints={}, lr=0.01):
#{{{
        """
        Stochatic gradient descent.
        """
        updates = []
        
        lr = theano.shared(np.float32(lr).astype(floatX))
        gradients = self.get_gradients(cost, params)
        
        for p, g in zip(params, gradients):
            v=-lr*g;
            new_p=p+v;
            # apply constraints
            if p in constraints:
                c=constraints[p];
                new_p=c(new_p);
            updates.append((p, new_p))

        return updates
#}}} 
Example 8
Project: Att-ChemdNER   Author: lingluodlut   File: optimization.py    License: Apache License 2.0 6 votes vote down vote up
def sgdmomentum(self, cost, params,constraints={}, lr=0.01,consider_constant=None, momentum=0.):
        """
        Stochatic gradient descent with momentum. Momentum has to be in [0, 1)
        """
        # Check that the momentum is a correct value
        assert 0 <= momentum < 1

        lr = theano.shared(np.float32(lr).astype(floatX))
        momentum = theano.shared(np.float32(momentum).astype(floatX))

        gradients = self.get_gradients(cost, params)
        velocities = [theano.shared(np.zeros_like(param.get_value(borrow=True)).astype(floatX)) for param in params]

        updates = []
        for param, gradient, velocity in zip(params, gradients, velocities):
            new_velocity = momentum * velocity - lr * gradient
            updates.append((velocity, new_velocity))
            new_p=param+new_velocity;
            # apply constraints
            if param in constraints:
                c=constraints[param];
                new_p=c(new_p);
            updates.append((param, new_p))
        return updates 
Example 9
Project: Att-ChemdNER   Author: lingluodlut   File: optimization.py    License: Apache License 2.0 6 votes vote down vote up
def adagrad(self, cost, params, lr=1.0, epsilon=1e-6,consider_constant=None):
        """
        Adagrad. Based on http://www.ark.cs.cmu.edu/cdyer/adagrad.pdf
        """
        lr = theano.shared(np.float32(lr).astype(floatX))
        epsilon = theano.shared(np.float32(epsilon).astype(floatX))

        gradients = self.get_gradients(cost, params,consider_constant)
        gsums = [theano.shared(np.zeros_like(param.get_value(borrow=True)).astype(floatX)) for param in params]

        updates = []
        for param, gradient, gsum in zip(params, gradients, gsums):
            new_gsum = gsum + gradient ** 2.
            updates.append((gsum, new_gsum))
            updates.append((param, param - lr * gradient / (T.sqrt(gsum + epsilon))))
        return updates 
Example 10
Project: Att-ChemdNER   Author: lingluodlut   File: optimization.py    License: Apache License 2.0 6 votes vote down vote up
def adadelta(self, cost, params, rho=0.95, epsilon=1e-6,consider_constant=None):
        """
        Adadelta. Based on:
        http://www.matthewzeiler.com/pubs/googleTR2012/googleTR2012.pdf
        """
        rho = theano.shared(np.float32(rho).astype(floatX))
        epsilon = theano.shared(np.float32(epsilon).astype(floatX))

        gradients = self.get_gradients(cost, params,consider_constant)
        accu_gradients = [theano.shared(np.zeros_like(param.get_value(borrow=True)).astype(floatX)) for param in params]
        accu_deltas = [theano.shared(np.zeros_like(param.get_value(borrow=True)).astype(floatX)) for param in params]

        updates = []
        for param, gradient, accu_gradient, accu_delta in zip(params, gradients, accu_gradients, accu_deltas):
            new_accu_gradient = rho * accu_gradient + (1. - rho) * gradient ** 2.
            delta_x = - T.sqrt((accu_delta + epsilon) / (new_accu_gradient + epsilon)) * gradient
            new_accu_delta = rho * accu_delta + (1. - rho) * delta_x ** 2.
            updates.append((accu_gradient, new_accu_gradient))
            updates.append((accu_delta, new_accu_delta))
            updates.append((param, param + delta_x))
        return updates 
Example 11
Project: Att-ChemdNER   Author: lingluodlut   File: optimization.py    License: Apache License 2.0 6 votes vote down vote up
def rmsprop(self, cost, params, lr=0.001, rho=0.9, eps=1e-6,consider_constant=None):
        """
        RMSProp.
        """
        lr = theano.shared(np.float32(lr).astype(floatX))

        gradients = self.get_gradients(cost, params,consider_constant)
        accumulators = [theano.shared(np.zeros_like(p.get_value()).astype(np.float32)) for p in params]

        updates = []

        for param, gradient, accumulator in zip(params, gradients, accumulators):
            new_accumulator = rho * accumulator + (1 - rho) * gradient ** 2
            updates.append((accumulator, new_accumulator))

            new_param = param - lr * gradient / T.sqrt(new_accumulator + eps)
            updates.append((param, new_param))

        return updates 
Example 12
Project: Att-ChemdNER   Author: lingluodlut   File: theano_backend.py    License: Apache License 2.0 6 votes vote down vote up
def in_top_k(predictions, targets, k):
    '''Returns whether the `targets` are in the top `k` `predictions`

    # Arguments
        predictions: A tensor of shape batch_size x classess and type float32.
        targets: A tensor of shape batch_size and type int32 or int64.
        k: An int, number of top elements to consider.

    # Returns
        A tensor of shape batch_size and type int. output_i is 1 if
        targets_i is within top-k values of predictions_i
    '''
    predictions_top_k = T.argsort(predictions)[:, -k:]
    result, _ = theano.map(lambda prediction, target: any(equal(prediction, target)), sequences=[predictions_top_k, targets])
    return result


# CONVOLUTIONS 
Example 13
Project: Att-ChemdNER   Author: lingluodlut   File: theano_backend.py    License: Apache License 2.0 6 votes vote down vote up
def ctc_path_probs(predict, Y, alpha=1e-4):
    smoothed_predict = (1 - alpha) * predict[:, Y] + alpha * np.float32(1.) / Y.shape[0]
    L = T.log(smoothed_predict)
    zeros = T.zeros_like(L[0])
    log_first = zeros

    f_skip_idxs = ctc_create_skip_idxs(Y)
    b_skip_idxs = ctc_create_skip_idxs(Y[::-1])  # there should be a shortcut to calculating this

    def step(log_f_curr, log_b_curr, f_active, log_f_prev, b_active, log_b_prev):
        f_active_next, log_f_next = ctc_update_log_p(f_skip_idxs, zeros, f_active, log_f_curr, log_f_prev)
        b_active_next, log_b_next = ctc_update_log_p(b_skip_idxs, zeros, b_active, log_b_curr, log_b_prev)
        return f_active_next, log_f_next, b_active_next, log_b_next

    [f_active, log_f_probs, b_active, log_b_probs], _ = theano.scan(
        step, sequences=[L, L[::-1, ::-1]], outputs_info=[np.int32(1), log_first, np.int32(1), log_first])

    idxs = T.arange(L.shape[1]).dimshuffle('x', 0)
    mask = (idxs < f_active.dimshuffle(0, 'x')) & (idxs < b_active.dimshuffle(0, 'x'))[::-1, ::-1]
    log_probs = log_f_probs + log_b_probs[::-1, ::-1] - L
    return log_probs, mask 
Example 14
def _get_rois_blob(im_rois, im_scale_factors):
    """Converts RoIs into network inputs.
    Arguments:
        im_rois (ndarray): R x 4 matrix of RoIs in original image coordinates
        im_scale_factors (list): scale factors as returned by _get_image_blob
    Returns:
        blob (ndarray): R x 5 matrix of RoIs in the image pyramid
    """
    rois_blob_real = []

    for i in range(len(im_scale_factors)):
        rois, levels = _project_im_rois(im_rois, np.array([im_scale_factors[i]]))
        rois_blob = np.hstack((levels, rois))
        rois_blob_real.append(rois_blob.astype(np.float32, copy=False))

    return rois_blob_real 
Example 15
def _get_rois_blob(im_rois, im_scale_factors):
    """Converts RoIs into network inputs.
    Arguments:
        im_rois (ndarray): R x 4 matrix of RoIs in original image coordinates
        im_scale_factors (list): scale factors as returned by _get_image_blob
    Returns:
        blob (ndarray): R x 5 matrix of RoIs in the image pyramid
    """
    rois_blob_real = []

    for i in range(len(im_scale_factors)):
        rois, levels = _project_im_rois(im_rois, np.array([im_scale_factors[i]]))
        rois_blob = np.hstack((levels, rois))
        rois_blob_real.append(rois_blob.astype(np.float32, copy=False))

    return rois_blob_real 
Example 16
def generate_anchors_pre(height, width, feat_stride, anchor_scales=(8,16,32), anchor_ratios=(0.5,1,2)):
  """ A wrapper function to generate anchors given different scales
    Also return the number of anchors in variable 'length'
  """
  anchors = generate_anchors(ratios=np.array(anchor_ratios), scales=np.array(anchor_scales))
  A = anchors.shape[0]
  shift_x = np.arange(0, width) * feat_stride
  shift_y = np.arange(0, height) * feat_stride
  shift_x, shift_y = np.meshgrid(shift_x, shift_y)
  shifts = np.vstack((shift_x.ravel(), shift_y.ravel(), shift_x.ravel(), shift_y.ravel())).transpose()
  K = shifts.shape[0]
  # width changes faster, so here it is H, W, C
  anchors = anchors.reshape((1, A, 4)) + shifts.reshape((1, K, 4)).transpose((1, 0, 2))
  anchors = anchors.reshape((K * A, 4)).astype(np.float32, copy=False)
  length = np.int32(anchors.shape[0])

  return anchors, length 
Example 17
Project: DDPAE-video-prediction   Author: jthsieh   File: moving_mnist.py    License: MIT License 6 votes vote down vote up
def generate_moving_mnist(self, num_digits=2):
    '''
    Get random trajectories for the digits and generate a video.
    '''
    data = np.zeros((self.n_frames_total, self.image_size_, self.image_size_), dtype=np.float32)
    for n in range(num_digits):
      # Trajectory
      start_y, start_x = self.get_random_trajectory(self.n_frames_total)
      ind = random.randint(0, self.mnist.shape[0] - 1)
      digit_image = self.mnist[ind]
      for i in range(self.n_frames_total):
        top    = start_y[i]
        left   = start_x[i]
        bottom = top + self.digit_size_
        right  = left + self.digit_size_
        # Draw digit
        data[i, top:bottom, left:right] = np.maximum(data[i, top:bottom, left:right], digit_image)

    data = data[..., np.newaxis]
    return data 
Example 18
Project: gated-graph-transformer-network   Author: hexahedria   File: ggtnn_train.py    License: MIT License 6 votes vote down vote up
def convert_answer(answer, num_words, format_spec, maxlen):
    """
    Convert an answer into an appropriate answer matrix given
    a ModelOutputFormat.

    num_words should be after processing with get_effective_answer_words,
    so that the last word is the "stop" word
    """
    assert format_spec in model.ModelOutputFormat
    if format_spec == model.ModelOutputFormat.subset:
        ans_mat = np.zeros((1,num_words), np.float32)
        for word in answer:
            ans_mat[0, word] = 1.0
    elif format_spec == model.ModelOutputFormat.category:
        ans_mat = np.zeros((1,num_words), np.float32)
        ans_mat[0,answer[0]] = 1.0
    elif format_spec == model.ModelOutputFormat.sequence:
        ans_mat = np.zeros((maxlen+1,num_words), np.float32)
        for i,word in enumerate(answer+[num_words-1]*(maxlen+1-len(answer))):
            ans_mat[i, word] = 1.0
    return ans_mat 
Example 19
Project: Traffic_sign_detection_YOLO   Author: AmeyaWagh   File: baseop.py    License: MIT License 6 votes vote down vote up
def wrap_variable(self, var):
        """wrap layer.w into variables"""
        val = self.lay.w.get(var, None)
        if val is None:
            shape = self.lay.wshape[var]
            args = [0., 1e-2, shape]
            if 'moving_mean' in var:
                val = np.zeros(shape)
            elif 'moving_variance' in var:
                val = np.ones(shape)
            else:
                val = np.random.normal(*args)
            self.lay.w[var] = val.astype(np.float32)
            self.act = 'Init '
        if not self.var: return

        val = self.lay.w[var]
        self.lay.w[var] = tf.constant_initializer(val)
        if var in self._SLIM: return
        with tf.variable_scope(self.scope):
            self.lay.w[var] = tf.get_variable(var,
                shape = self.lay.wshape[var],
                dtype = tf.float32,
                initializer = self.lay.w[var]) 
Example 20
Project: kaldi-python-io   Author: funcwj   File: _io_kernel.py    License: Apache License 2.0 6 votes vote down vote up
def read_common_mat(fd):
    """ 
        Read common matrix(for class Matrix in kaldi setup)
        see matrix/kaldi-matrix.cc::
            void Matrix<Real>::Read(std::istream & is, bool binary, bool add)
        Return a numpy ndarray object
    """
    mat_type = read_token(fd)
    print_info(f'\tType of the common matrix: {mat_type}')
    if mat_type not in ["FM", "DM"]:
        raise RuntimeError(f"Unknown matrix type in kaldi: {mat_type}")
    float_size = 4 if mat_type == 'FM' else 8
    float_type = np.float32 if mat_type == 'FM' else np.float64
    num_rows = read_int32(fd)
    num_cols = read_int32(fd)
    print_info(f'\tSize of the common matrix: {num_rows} x {num_cols}')
    mat_data = fd.read(float_size * num_cols * num_rows)
    mat = np.fromstring(mat_data, dtype=float_type)
    return mat.reshape(num_rows, num_cols) 
Example 21
Project: kaldi-python-io   Author: funcwj   File: _io_kernel.py    License: Apache License 2.0 6 votes vote down vote up
def read_float_vec(fd, direct_access=False):
    """
        Read float vector(for class Vector in kaldi setup)
        see matrix/kaldi-vector.cc
    """
    if direct_access:
        expect_binary(fd)
    vec_type = read_token(fd)
    print_info(f'\tType of the common vector: {vec_type}')
    if vec_type not in ["FV", "DV"]:
        raise RuntimeError(f"Unknown matrix type in kaldi: {vec_type}")
    float_size = 4 if vec_type == 'FV' else 8
    float_type = np.float32 if vec_type == 'FV' else np.float64
    dim = read_int32(fd)
    print_info(f'\tDim of the common vector: {dim}')
    vec_data = fd.read(float_size * dim)
    return np.fromstring(vec_data, dtype=float_type) 
Example 22
Project: disentangling_conditional_gans   Author: zalandoresearch   File: tfutil.py    License: MIT License 6 votes vote down vote up
def autosummary(name, value):
    id = name.replace('/', '_')
    if is_tf_expression(value):
        with tf.name_scope('summary_' + id), tf.device(value.device):
            update_op = _create_autosummary_var(name, value)
            with tf.control_dependencies([update_op]):
                return tf.identity(value)
    else: # python scalar or numpy array
        if name not in _autosummary_immediate:
            with absolute_name_scope('Autosummary/' + id), tf.device(None), tf.control_dependencies(None):
                update_value = tf.placeholder(tf.float32)
                update_op = _create_autosummary_var(name, update_value)
                _autosummary_immediate[name] = update_op, update_value
        update_op, update_value = _autosummary_immediate[name]
        run(update_op, {update_value: np.float32(value)})
        return value

# Create the necessary ops to include autosummaries in TensorBoard report.
# Note: This should be done only once per graph. 
Example 23
Project: disentangling_conditional_gans   Author: zalandoresearch   File: tfutil.py    License: MIT License 6 votes vote down vote up
def _create_autosummary_var(name, value_expr):
    assert not _autosummary_finalized
    v = tf.cast(value_expr, tf.float32)
    if v.shape.ndims is 0:
        v = [v, np.float32(1.0)]
    elif v.shape.ndims is 1:
        v = [tf.reduce_sum(v), tf.cast(tf.shape(v)[0], tf.float32)]
    else:
        v = [tf.reduce_sum(v), tf.reduce_prod(tf.cast(tf.shape(v), tf.float32))]
    v = tf.cond(tf.is_finite(v[0]), lambda: tf.stack(v), lambda: tf.zeros(2))
    with tf.control_dependencies(None):
        var = tf.Variable(tf.zeros(2)) # [numerator, denominator]
    update_op = tf.cond(tf.is_variable_initialized(var), lambda: tf.assign_add(var, v), lambda: tf.assign(var, v))
    if name in _autosummary_vars:
        _autosummary_vars[name].append(var)
    else:
        _autosummary_vars[name] = [var]
    return update_op

#----------------------------------------------------------------------------
# Call filewriter.add_summary() with all summaries in the default graph,
# automatically finalizing and merging them on the first call. 
Example 24
Project: disentangling_conditional_gans   Author: zalandoresearch   File: networks.py    License: MIT License 6 votes vote down vote up
def minibatch_stddev_layer(x, group_size=4):
    with tf.variable_scope('MinibatchStddev'):
        group_size = tf.minimum(group_size, tf.shape(x)[0])     # Minibatch must be divisible by (or smaller than) group_size.
        s = x.shape                                             # [NCHW]  Input shape.
        y = tf.reshape(x, [group_size, -1, s[1], s[2], s[3]])   # [GMCHW] Split minibatch into M groups of size G.
        y = tf.cast(y, tf.float32)                              # [GMCHW] Cast to FP32.
        y -= tf.reduce_mean(y, axis=0, keep_dims=True)           # [GMCHW] Subtract mean over group.
        y = tf.reduce_mean(tf.square(y), axis=0)                # [MCHW]  Calc variance over group.
        y = tf.sqrt(y + 1e-8)                                   # [MCHW]  Calc stddev over group.
        y = tf.reduce_mean(y, axis=[1,2,3], keep_dims=True)      # [M111]  Take average over fmaps and pixels.
        y = tf.cast(y, x.dtype)                                 # [M111]  Cast back to original data type.
        y = tf.tile(y, [group_size, 1, s[2], s[3]])             # [N1HW]  Replicate over group and pixels.
        return tf.concat([x, y], axis=1)                        # [NCHW]  Append as new fmap.

#----------------------------------------------------------------------------
# Generator network used in the paper. 
Example 25
Project: disentangling_conditional_gans   Author: zalandoresearch   File: dataset_tool.py    License: MIT License 6 votes vote down vote up
def add_image(self, img):
        if self.print_progress and self.cur_images % self.progress_interval == 0:
            print('%d / %d\r' % (self.cur_images, self.expected_images), end='', flush=True)
            sys.stdout.flush()
        if self.shape is None:
            self.shape = img.shape
            self.resolution_log2 = int(np.log2(self.shape[1]))
            assert self.shape[0] in [1, 3]
            assert self.shape[1] == self.shape[2]
            assert self.shape[1] == 2**self.resolution_log2
            tfr_opt = tf.python_io.TFRecordOptions(tf.python_io.TFRecordCompressionType.NONE)
            for lod in range(self.resolution_log2 - 1):
                tfr_file = self.tfr_prefix + '-r%02d.tfrecords' % (self.resolution_log2 - lod)
                self.tfr_writers.append(tf.python_io.TFRecordWriter(tfr_file, tfr_opt))
        assert img.shape == self.shape
        for lod, tfr_writer in enumerate(self.tfr_writers):
            if lod:
                img = img.astype(np.float32)
                img = (img[:, 0::2, 0::2] + img[:, 0::2, 1::2] + img[:, 1::2, 0::2] + img[:, 1::2, 1::2]) * 0.25
            quant = np.rint(img).clip(0, 255).astype(np.uint8)
            ex = tf.train.Example(features=tf.train.Features(feature={
                'shape': tf.train.Feature(int64_list=tf.train.Int64List(value=quant.shape)),
                'data': tf.train.Feature(bytes_list=tf.train.BytesList(value=[quant.tostring()]))}))
            tfr_writer.write(ex.SerializeToString())
        self.cur_images += 1 
Example 26
Project: disentangling_conditional_gans   Author: zalandoresearch   File: dataset_tool.py    License: MIT License 6 votes vote down vote up
def create_mnist(tfrecord_dir, mnist_dir):
    print('Loading MNIST from "%s"' % mnist_dir)
    import gzip
    with gzip.open(os.path.join(mnist_dir, 'train-images-idx3-ubyte.gz'), 'rb') as file:
        images = np.frombuffer(file.read(), np.uint8, offset=16)
    with gzip.open(os.path.join(mnist_dir, 'train-labels-idx1-ubyte.gz'), 'rb') as file:
        labels = np.frombuffer(file.read(), np.uint8, offset=8)
    images = images.reshape(-1, 1, 28, 28)
    images = np.pad(images, [(0,0), (0,0), (2,2), (2,2)], 'constant', constant_values=0)
    assert images.shape == (60000, 1, 32, 32) and images.dtype == np.uint8
    assert labels.shape == (60000,) and labels.dtype == np.uint8
    assert np.min(images) == 0 and np.max(images) == 255
    assert np.min(labels) == 0 and np.max(labels) == 9
    onehot = np.zeros((labels.size, np.max(labels) + 1), dtype=np.float32)
    onehot[np.arange(labels.size), labels] = 1.0
    
    with TFRecordExporter(tfrecord_dir, images.shape[0]) as tfr:
        order = tfr.choose_shuffled_order()
        for idx in range(order.size):
            tfr.add_image(images[order[idx]])
        tfr.add_labels(onehot[order])

#---------------------------------------------------------------------------- 
Example 27
Project: disentangling_conditional_gans   Author: zalandoresearch   File: dataset_tool.py    License: MIT License 6 votes vote down vote up
def create_cifar100(tfrecord_dir, cifar100_dir):
    print('Loading CIFAR-100 from "%s"' % cifar100_dir)
    import pickle
    with open(os.path.join(cifar100_dir, 'train'), 'rb') as file:
        data = pickle.load(file, encoding='latin1')
    images = data['data'].reshape(-1, 3, 32, 32)
    labels = np.array(data['fine_labels'])
    assert images.shape == (50000, 3, 32, 32) and images.dtype == np.uint8
    assert labels.shape == (50000,) and labels.dtype == np.int32
    assert np.min(images) == 0 and np.max(images) == 255
    assert np.min(labels) == 0 and np.max(labels) == 99
    onehot = np.zeros((labels.size, np.max(labels) + 1), dtype=np.float32)
    onehot[np.arange(labels.size), labels] = 1.0

    with TFRecordExporter(tfrecord_dir, images.shape[0]) as tfr:
        order = tfr.choose_shuffled_order()
        for idx in range(order.size):
            tfr.add_image(images[order[idx]])
        tfr.add_labels(onehot[order])

#---------------------------------------------------------------------------- 
Example 28
Project: disentangling_conditional_gans   Author: zalandoresearch   File: util_scripts.py    License: MIT License 6 votes vote down vote up
def generate_fake_images(run_id, snapshot=None, grid_size=[1,1], num_pngs=1, image_shrink=1, png_prefix=None, random_seed=1000, minibatch_size=8):
    network_pkl = misc.locate_network_pkl(run_id, snapshot)
    if png_prefix is None:
        png_prefix = misc.get_id_string_for_network_pkl(network_pkl) + '-'
    random_state = np.random.RandomState(random_seed)

    print('Loading network from "%s"...' % network_pkl)
    G, D, Gs = misc.load_network_pkl(run_id, snapshot)

    result_subdir = misc.create_result_subdir(config.result_dir, config.desc)
    for png_idx in range(num_pngs):
        print('Generating png %d / %d...' % (png_idx, num_pngs))
        latents = misc.random_latents(np.prod(grid_size), Gs, random_state=random_state)
        labels = np.zeros([latents.shape[0], 0], np.float32)
        images = Gs.run(latents, labels, minibatch_size=minibatch_size, num_gpus=config.num_gpus, out_mul=127.5, out_add=127.5, out_shrink=image_shrink, out_dtype=np.uint8)
        misc.save_image_grid(images, os.path.join(result_subdir, '%s%06d.png' % (png_prefix, png_idx)), [0,255], grid_size)
    open(os.path.join(result_subdir, '_done.txt'), 'wt').close()

#----------------------------------------------------------------------------
# Generate MP4 video of random interpolations using a previously trained network.
# To run, uncomment the appropriate line in config.py and launch train.py. 
Example 29
Project: Black-Box-Audio   Author: rtaori   File: run_audio_attack.py    License: MIT License 5 votes vote down vote up
def setup_graph(self, input_audio_batch, target_phrase): 
        batch_size = input_audio_batch.shape[0]
        weird = (input_audio_batch.shape[1] - 1) // 320 
        logits_arg2 = np.tile(weird, batch_size)
        dense_arg1 = np.array(np.tile(target_phrase, (batch_size, 1)), dtype=np.int32)
        dense_arg2 = np.array(np.tile(target_phrase.shape[0], batch_size), dtype=np.int32)
        
        pass_in = np.clip(input_audio_batch, -2**15, 2**15-1)
        seq_len = np.tile(weird, batch_size).astype(np.int32)
        
        with tf.variable_scope('', reuse=tf.AUTO_REUSE):
            
            inputs = tf.placeholder(tf.float32, shape=pass_in.shape, name='a')
            len_batch = tf.placeholder(tf.float32, name='b')
            arg2_logits = tf.placeholder(tf.int32, shape=logits_arg2.shape, name='c')
            arg1_dense = tf.placeholder(tf.float32, shape=dense_arg1.shape, name='d')
            arg2_dense = tf.placeholder(tf.int32, shape=dense_arg2.shape, name='e')
            len_seq = tf.placeholder(tf.int32, shape=seq_len.shape, name='f')
            
            logits = get_logits(inputs, arg2_logits)
            target = ctc_label_dense_to_sparse(arg1_dense, arg2_dense, len_batch)
            ctcloss = tf.nn.ctc_loss(labels=tf.cast(target, tf.int32), inputs=logits, sequence_length=len_seq)
            decoded, _ = tf.nn.ctc_greedy_decoder(logits, arg2_logits, merge_repeated=True)
            
            sess = tf.Session()
            saver = tf.train.Saver(tf.global_variables())
            saver.restore(sess, "models/session_dump")
            
        func1 = lambda a, b, c, d, e, f: sess.run(ctcloss, 
            feed_dict={inputs: a, len_batch: b, arg2_logits: c, arg1_dense: d, arg2_dense: e, len_seq: f})
        func2 = lambda a, b, c, d, e, f: sess.run([ctcloss, decoded], 
            feed_dict={inputs: a, len_batch: b, arg2_logits: c, arg1_dense: d, arg2_dense: e, len_seq: f})
        return (func1, func2) 
Example 30
Project: Black-Box-Audio   Author: rtaori   File: tf_logits.py    License: MIT License 5 votes vote down vote up
def compute_mfcc(audio, **kwargs):
    """
    Compute the MFCC for a given audio waveform. This is
    identical to how DeepSpeech does it, but does it all in
    TensorFlow so that we can differentiate through it.
    """

    batch_size, size = audio.get_shape().as_list()
    audio = tf.cast(audio, tf.float32)

    # 1. Pre-emphasizer, a high-pass filter
    audio = tf.concat((audio[:, :1], audio[:, 1:] - 0.97*audio[:, :-1], np.zeros((batch_size,1000),dtype=np.float32)), 1)

    # 2. windowing into frames of 320 samples, overlapping
    windowed = tf.stack([audio[:, i:i+400] for i in range(0,size-320,160)],1)

    # 3. Take the FFT to convert to frequency space
    ffted = tf.spectral.rfft(windowed, [512])
    ffted = 1.0 / 512 * tf.square(tf.abs(ffted))

    # 4. Compute the Mel windowing of the FFT
    energy = tf.reduce_sum(ffted,axis=2)+1e-30
    filters = np.load("filterbanks.npy").T
    feat = tf.matmul(ffted, np.array([filters]*batch_size,dtype=np.float32))+1e-30

    # 5. Take the DCT again, because why not
    feat = tf.log(feat)
    feat = tf.spectral.dct(feat, type=2, norm='ortho')[:,:,:26]

    # 6. Amplify high frequencies for some reason
    _,nframes,ncoeff = feat.get_shape().as_list()
    n = np.arange(ncoeff)
    lift = 1 + (22/2.)*np.sin(np.pi*n/22)
    feat = lift*feat
    width = feat.get_shape().as_list()[1]

    # 7. And now stick the energy next to the features
    feat = tf.concat((tf.reshape(tf.log(energy),(-1,width,1)), feat[:, :, 1:]), axis=2)
    
    return feat