Python tensorflow.fill() Examples

def soften_labels(bool_labels, softness=0.05, scope='soften_labels'):
  """Converts boolean labels into float32.

  Args:
    bool_labels: Tensor with dtype `boolean`
    softness: The float value to use for False.  1 - softness is implicitly used
              for True
    scope: passed to op_scope

  Returns:
    Tensor with same shape as bool_labels with dtype `float32` and values 0.05
    for False and 0.95 for True.
  """
  with tf.op_scope([bool_labels, softness], scope):
    label_shape = tf.shape(bool_labels, name='label_shape')
    return tf.where(bool_labels,
                    tf.fill(label_shape, 1.0 - softness, name='soft_true'),
                    tf.fill(label_shape, softness, name='soft_false')) 

def decode(self, inputs, from_pred=True, method='greedy'):
        if from_pred:
            logit_length = tf.fill([tf.shape(inputs)[0]], tf.shape(inputs)[1])
            if method == 'greedy':
                decoded, _ = tf.nn.ctc_greedy_decoder(
                    inputs=tf.transpose(inputs, perm=[1, 0, 2]),
                    sequence_length=logit_length,
                    merge_repeated=self.merge_repeated)
            elif method == 'beam_search':
                decoded, _ = tf.nn.ctc_beam_search_decoder(
                    inputs=tf.transpose(inputs, perm=[1, 0, 2]),
                    sequence_length=logit_length)
            inputs = decoded[0]
        decoded = tf.sparse.to_dense(inputs, 
                                     default_value=self.blank_index).numpy()
        decoded = self.map2string(decoded)
        return decoded 

def create_model(self, model_input, vocab_size, num_frames, l2_penalty=1e-8, **unused_params):
    """
    A super model that combine one or more models
    """
    models = FLAGS.wide_and_deep_models
    outputs = []
    for model_name in map(lambda x: x.strip(), models.split(",")):
      model = getattr(frame_level_models, model_name, None)()
      output = model.create_model(model_input, vocab_size, num_frames, l2_penalty=l2_penalty, **unused_params)["predictions"]
      outputs.append(tf.expand_dims(output, axis=2))
    num_models = len(outputs)
    model_outputs = tf.concat(outputs, axis=2)
#    linear_combination = tf.get_variable("combine", shape=[vocab_size,num_models],
#        dtype=tf.float32, initializer=tf.zeros_initializer(),
#        regularizer=slim.l2_regularizer(l2_penalty))
#    combination = tf.nn.softmax(linear_combination)
    combination = tf.fill(dims=[vocab_size,num_models], value=1.0/num_models)
    output_sum = tf.einsum("ijk,jk->ij", model_outputs, combination)
    return {"predictions": output_sum} 

def _convert_inputs(self, inputs):
        # Convert from time-major to batch-major. Note that we take factor 0
        # from x and ignore any other factors.
        source_ids = tf.transpose(a=inputs.x[0], perm=[1,0])
        source_mask = tf.transpose(a=inputs.x_mask, perm=[1,0])
        target_ids_out = tf.transpose(a=inputs.y, perm=[1,0])
        target_mask = tf.transpose(a=inputs.y_mask, perm=[1,0])

        # target_ids_in is a bit more complicated since we need to insert
        # the special <GO> symbol (with value 1) at the start of each sentence
        max_len, batch_size = tf.shape(input=inputs.y)[0], tf.shape(input=inputs.y)[1]
        go_symbols = tf.fill(value=1, dims=[1, batch_size])
        tmp = tf.concat([go_symbols, inputs.y], 0)
        tmp = tmp[:-1, :]
        target_ids_in = tf.transpose(a=tmp, perm=[1,0])
        return (source_ids, source_mask, target_ids_in, target_ids_out,
                target_mask) 

def testRandomPixelValueScale(self):
    preprocessing_options = []
    preprocessing_options.append((preprocessor.normalize_image, {
        'original_minval': 0,
        'original_maxval': 255,
        'target_minval': 0,
        'target_maxval': 1
    }))
    preprocessing_options.append((preprocessor.random_pixel_value_scale, {}))
    images = self.createTestImages()
    tensor_dict = {fields.InputDataFields.image: images}
    tensor_dict = preprocessor.preprocess(tensor_dict, preprocessing_options)
    images_min = tf.to_float(images) * 0.9 / 255.0
    images_max = tf.to_float(images) * 1.1 / 255.0
    images = tensor_dict[fields.InputDataFields.image]
    values_greater = tf.greater_equal(images, images_min)
    values_less = tf.less_equal(images, images_max)
    values_true = tf.fill([1, 4, 4, 3], True)
    with self.test_session() as sess:
      (values_greater_, values_less_, values_true_) = sess.run(
          [values_greater, values_less, values_true])
      self.assertAllClose(values_greater_, values_true_)
      self.assertAllClose(values_less_, values_true_) 

def simulate(self, action):
    with tf.name_scope("environment/simulate"):  # Do we need this?
      initializer = (tf.zeros_like(self._observ),
                     tf.fill((len(self),), 0.0), tf.fill((len(self),), False))

      def not_done_step(a, _):
        reward, done = self._batch_env.simulate(action)
        with tf.control_dependencies([reward, done]):
          # TODO(piotrmilos): possibly ignore envs with done
          r0 = tf.maximum(a[0], self._batch_env.observ)
          r1 = tf.add(a[1], reward)
          r2 = tf.logical_or(a[2], done)

          return (r0, r1, r2)

      simulate_ret = tf.scan(not_done_step, tf.range(self.skip),
                             initializer=initializer, parallel_iterations=1,
                             infer_shape=False)
      simulate_ret = [ret[-1, ...] for ret in simulate_ret]

      with tf.control_dependencies([self._observ.assign(simulate_ret[0])]):
        return tf.identity(simulate_ret[1]), tf.identity(simulate_ret[2]) 

def simulate(self, action):
    with tf.name_scope("environment/simulate"):  # Do we need this?
      initializer = (tf.zeros(self.old_shape, dtype=tf.float32),
                     tf.fill((len(self),), 0.0), tf.fill((len(self),), False))

      def not_done_step(a, _):
        reward, done = self._batch_env.simulate(action)
        with tf.control_dependencies([reward, done]):
          r0 = self._batch_env.observ + 0
          r1 = tf.add(a[1], reward)
          r2 = tf.logical_or(a[2], done)
          return (r0, r1, r2)

      simulate_ret = tf.scan(not_done_step, tf.range(self.skip),
                             initializer=initializer, parallel_iterations=1,
                             infer_shape=False)
      observations, rewards, dones = simulate_ret
      split_observations = tf.split(observations, self.skip, axis=0)
      split_observations = [tf.squeeze(o, axis=0) for o in split_observations]
      observation = tf.concat(split_observations, axis=-1)
      with tf.control_dependencies([self._observ.assign(observation)]):
        return tf.identity(rewards[-1, ...]), tf.identity(dones[-1, ...]) 

def resize_axis(tensor, axis, new_size, fill_value=0):
  tensor = tf.convert_to_tensor(tensor)
  shape = tf.unstack(tf.shape(tensor))

  pad_shape = shape[:]
  pad_shape[axis] = tf.maximum(0, new_size - shape[axis])

  shape[axis] = tf.minimum(shape[axis], new_size)
  shape = tf.stack(shape)

  resized = tf.concat([
      tf.slice(tensor, tf.zeros_like(shape), shape),
      tf.fill(tf.stack(pad_shape), tf.cast(fill_value, tensor.dtype))
  ], axis)

  # Update shape.
  new_shape = tensor.get_shape().as_list()  # A copy is being made.
  new_shape[axis] = new_size
  resized.set_shape(new_shape)
  return resized 

def update_state(self, y_true, y_pred, sample_weight=None):
        """
        Maybe have more fast implementation.
        """
        b = tf.shape(y_true)[0]
        max_width = tf.maximum(tf.shape(y_true)[1], tf.shape(y_pred)[1])
        logit_length = tf.fill([tf.shape(y_pred)[0]], tf.shape(y_pred)[1])        
        decoded, _ = tf.nn.ctc_greedy_decoder(
            inputs=tf.transpose(y_pred, perm=[1, 0, 2]),
            sequence_length=logit_length)
        y_true = tf.sparse.reset_shape(y_true, [b, max_width])
        y_pred = tf.sparse.reset_shape(decoded[0], [b, max_width])
        y_true = tf.sparse.to_dense(y_true, default_value=-1)
        y_pred = tf.sparse.to_dense(y_pred, default_value=-1)
        y_true = tf.cast(y_true, tf.int32)
        y_pred = tf.cast(y_pred, tf.int32)
        values = tf.math.reduce_any(tf.math.not_equal(y_true, y_pred), axis=1)
        values = tf.cast(values, tf.int32)
        values = tf.reduce_sum(values)
        self.total.assign_add(b)
        self.count.assign_add(b - values) 

def testRandomPixelValueScale(self):
    preprocessing_options = []
    preprocessing_options.append((preprocessor.normalize_image, {
        'original_minval': 0,
        'original_maxval': 255,
        'target_minval': 0,
        'target_maxval': 1
    }))
    preprocessing_options.append((preprocessor.random_pixel_value_scale, {}))
    images = self.createTestImages()
    tensor_dict = {fields.InputDataFields.image: images}
    tensor_dict = preprocessor.preprocess(tensor_dict, preprocessing_options)
    images_min = tf.to_float(images) * 0.9 / 255.0
    images_max = tf.to_float(images) * 1.1 / 255.0
    images = tensor_dict[fields.InputDataFields.image]
    values_greater = tf.greater_equal(images, images_min)
    values_less = tf.less_equal(images, images_max)
    values_true = tf.fill([1, 4, 4, 3], True)
    with self.test_session() as sess:
      (values_greater_, values_less_, values_true_) = sess.run(
          [values_greater, values_less, values_true])
      self.assertAllClose(values_greater_, values_true_)
      self.assertAllClose(values_less_, values_true_) 

def testRandomPixelValueScale(self):
    preprocessing_options = []
    preprocessing_options.append((preprocessor.normalize_image, {
        'original_minval': 0,
        'original_maxval': 255,
        'target_minval': 0,
        'target_maxval': 1
    }))
    preprocessing_options.append((preprocessor.random_pixel_value_scale, {}))
    images = self.createTestImages()
    tensor_dict = {fields.InputDataFields.image: images}
    tensor_dict = preprocessor.preprocess(tensor_dict, preprocessing_options)
    images_min = tf.to_float(images) * 0.9 / 255.0
    images_max = tf.to_float(images) * 1.1 / 255.0
    images = tensor_dict[fields.InputDataFields.image]
    values_greater = tf.greater_equal(images, images_min)
    values_less = tf.less_equal(images, images_max)
    values_true = tf.fill([1, 4, 4, 3], True)
    with self.test_session() as sess:
      (values_greater_, values_less_, values_true_) = sess.run(
          [values_greater, values_less, values_true])
      self.assertAllClose(values_greater_, values_true_)
      self.assertAllClose(values_less_, values_true_) 

def initialize(self, dtype=tf.float64):
        if self.tf_mean is None:
            if self.mean is not None:
                self.tf_mean = tf.Variable(self.mean, dtype=dtype)
            else:
                self.tf_mean = tf.Variable(tf.cast(tf.fill([self.dims], 0.0), dtype))

        if self.tf_covariance is None:
            if self.covariance is not None:
                self.tf_covariance = self.covariance
            else:
                self.tf_covariance = FullCovariance(self.dims)

            self.tf_covariance.initialize(dtype)

        if self.tf_ln2piD is None:
            self.tf_ln2piD = tf.constant(np.log(2 * np.pi) * self.dims, dtype=dtype) 

def plot_fitted_data(points, c_means, c_variances):
    """Plots the data and given Gaussian components"""
    plt.plot(points[:, 0], points[:, 1], "b.", zorder=0)
    plt.plot(c_means[:, 0], c_means[:, 1], "r.", zorder=1)

    for i in range(c_means.shape[0]):
        std = np.sqrt(c_variances[i])
        plt.axes().add_artist(pat.Ellipse(
            c_means[i], 2 * std[0], 2 * std[1],
            fill=False, color="red", linewidth=2, zorder=1
        ))

    plt.show()


# PREPARING DATA

# generating DATA_POINTS points from a GMM with COMPONENTS components 

def UniformInt(name=None):
    X = tf.placeholder(config.int_dtype, name=name)

    Distribution.logp = tf.fill(tf.shape(X), config.dtype(0))

    def integral(lower, upper):
        val = tf.cond(
            tf.logical_or(
                tf.is_inf(tf.ceil(tf.cast(lower, config.dtype))),
                tf.is_inf(tf.floor(tf.cast(upper, config.dtype)))
            ),
            lambda: tf.constant(1, dtype=config.dtype),
            lambda: tf.cast(upper, config.dtype) - tf.cast(lower, config.dtype),
        )
        return val

    Distribution.integral = integral

    return X 

def testRandomPixelValueScale(self):
    preprocessing_options = []
    preprocessing_options.append((preprocessor.normalize_image, {
        'original_minval': 0,
        'original_maxval': 255,
        'target_minval': 0,
        'target_maxval': 1
    }))
    preprocessing_options.append((preprocessor.random_pixel_value_scale, {}))
    images = self.createTestImages()
    tensor_dict = {fields.InputDataFields.image: images}
    tensor_dict = preprocessor.preprocess(tensor_dict, preprocessing_options)
    images_min = tf.to_float(images) * 0.9 / 255.0
    images_max = tf.to_float(images) * 1.1 / 255.0
    images = tensor_dict[fields.InputDataFields.image]
    values_greater = tf.greater_equal(images, images_min)
    values_less = tf.less_equal(images, images_max)
    values_true = tf.fill([1, 4, 4, 3], True)
    with self.test_session() as sess:
      (values_greater_, values_less_, values_true_) = sess.run(
          [values_greater, values_less, values_true])
      self.assertAllClose(values_greater_, values_true_)
      self.assertAllClose(values_less_, values_true_) 

def _build(self, obs_input, name=None):
        """Build model.

        Args:
            obs_input (tf.Tensor): Entire time-series observation input.
            name (str): Inner model name, also the variable scope of the
                inner model, if exist. One example is
                garage.tf.models.Sequential.

        Returns:
            tf.tensor: Action.
            tf.tensor: Mean.
            tf.Tensor: Log of standard deviation.
            garage.distributions.DiagonalGaussian: Distribution.

        """
        del name
        return_var = tf.compat.v1.get_variable(
            'return_var', (), initializer=tf.constant_initializer(0.5))
        mean = tf.fill((tf.shape(obs_input)[0], self.output_dim), return_var)
        log_std = tf.fill((tf.shape(obs_input)[0], self.output_dim), 0.5)
        action = mean + log_std * 0.5
        dist = DiagonalGaussian(self.output_dim)
        # action will be 0.5 + 0.5 * 0.5 = 0.75
        return action, mean, log_std, dist 

def _build(self, obs_input, name=None):
        """Build model.

        Args:
            obs_input (tf.Tensor): Entire time-series observation input.
            name (str): Inner model name, also the variable scope of the
                inner model, if exist. One example is
                garage.tf.models.Sequential.

        Returns:
            tf.tensor: Mean.
            tf.Tensor: Log of standard deviation.
            garage.distributions.DiagonalGaussian: Distribution.

        """
        del name
        return_var = tf.compat.v1.get_variable(
            'return_var', (), initializer=tf.constant_initializer(0.5))
        mean = tf.fill((tf.shape(obs_input)[0], self.output_dim), return_var)
        log_std = tf.fill((tf.shape(obs_input)[0], self.output_dim),
                          np.log(0.5))
        dist = DiagonalGaussian(self.output_dim)
        # action will be 0.5 + 0.5 * 0.5 = 0.75
        return mean, log_std, dist 

def where(cond, true, false, name=None):
    """Similar to tf.where, but broadcasts scalar values."""
    with tf.name_scope(name, 'where', [cond, true, false]) as name:
        cond = tf.convert_to_tensor(cond, name='cond', dtype=tf.bool)
        true = tf.convert_to_tensor(true, name='true',
                                    dtype=false.dtype if isinstance(false, tf.Tensor) else None)
        false = tf.convert_to_tensor(false, name='false', dtype=true.dtype)
        if true.shape.rank == false.shape.rank == 0:
            shape = tf.shape(cond)
            true = tf.fill(shape, true)
            false = tf.fill(shape, false)
        elif true.shape.rank == 0:
            true = tf.fill(tf.shape(false), true)
        elif false.shape.rank == 0:
            false = tf.fill(tf.shape(true), false)
        return tf.where(cond, true, false, name=name) 

def _build(self, obs_input, name=None):
        """Build model given input placeholder(s).

        Args:
            obs_input (tf.Tensor): Tensor input for state.
            name (str): Inner model name, also the variable scope of the
                inner model, if exist. One example is
                garage.tf.models.Sequential.

        Return:
            tf.Tensor: Tensor output of the model.

        """
        del name
        return_var = tf.compat.v1.get_variable(
            'return_var', (), initializer=tf.constant_initializer(0.5))
        return tf.fill((tf.shape(obs_input)[0], self.output_dim), return_var) 

def _build(self, obs_input, act_input, name=None):
        """Build model given input placeholder(s).

        Args:
            obs_input (tf.Tensor): Tensor input for state.
            act_input (tf.Tensor): Tensor input for action.
            name (str): Inner model name, also the variable scope of the
                inner model, if exist. One example is
                garage.tf.models.Sequential.

        Return:
            tf.Tensor: Tensor output of the model.

        """
        del name
        del act_input
        return_var = tf.compat.v1.get_variable(
            'return_var', (), initializer=tf.constant_initializer(0.5))
        return tf.fill((tf.shape(obs_input)[0], self.output_dim), return_var) 

def testRandomPixelValueScale(self):
    preprocessing_options = []
    preprocessing_options.append((preprocessor.normalize_image, {
        'original_minval': 0,
        'original_maxval': 255,
        'target_minval': 0,
        'target_maxval': 1
    }))
    preprocessing_options.append((preprocessor.random_pixel_value_scale, {}))
    images = self.createTestImages()
    tensor_dict = {fields.InputDataFields.image: images}
    tensor_dict = preprocessor.preprocess(tensor_dict, preprocessing_options)
    images_min = tf.to_float(images) * 0.9 / 255.0
    images_max = tf.to_float(images) * 1.1 / 255.0
    images = tensor_dict[fields.InputDataFields.image]
    values_greater = tf.greater_equal(images, images_min)
    values_less = tf.less_equal(images, images_max)
    values_true = tf.fill([1, 4, 4, 3], True)
    with self.test_session() as sess:
      (values_greater_, values_less_, values_true_) = sess.run(
          [values_greater, values_less, values_true])
      self.assertAllClose(values_greater_, values_true_)
      self.assertAllClose(values_less_, values_true_) 

def version_1(cls, node, **kwargs):
    tensor_dict = kwargs["tensor_dict"]

    if "shape" in node.attrs:
      shape = node.attrs["shape"]
    else:
      shape = tensor_dict[
          node.inputs[0]].get_shape().as_list() if node.attrs.get(
              "input_as_shape", 0) == 0 else tensor_dict[node.inputs[0]]

    if "extra_shape" in node.attrs:
      shape = tf.concat([shape, node.attrs["extra_shape"]], 0)

    value = node.attrs.get("value", 0.)

    if "dtype" in node.attrs:
      return [tf.cast(tf.fill(shape, value), dtype=node.attrs["dtype"])]

    return [cls.make_tensor_from_onnx_node(node, inputs=[shape], **kwargs)] 

def get_multi_dataset(datasets, pmf=None):
  """Returns a Dataset that samples records from one or more Datasets.

  Args:
    datasets: A list of one or more Dataset objects to sample from.
    pmf: A tensor of shape [len(datasets)], the probabilities to sample each
      dataset with. This tensor is often constructed with the global_step. If
      this is None, we sample from the datasets uniformly at random.

  Returns:
    A Dataset object containing records from multiple datasets. Note that
    because this dataset iterates through other datasets it is stateful, thus
    you will need to call make_initializable_iterator instead of
    make_one_shot_iterator.
  """
  pmf = tf.fill([len(datasets)], 1.0 / len(datasets)) if pmf is None else pmf
  samplers = [d.repeat().make_one_shot_iterator().get_next for d in datasets]
  sample = lambda _: categorical_case(pmf, samplers)
  return tf.data.Dataset.from_tensors([]).repeat().map(sample) 

def reset(self, entries_to_reset):
    """Reset the entries in the memory.

    Args:
      entries_to_reset: a 1D tensor.
    Returns:
      the reset op.
    """
    num_updates = tf.size(entries_to_reset)
    update_vals = tf.scatter_update(
        self.mem_vals, entries_to_reset,
        tf.tile(tf.expand_dims(
            tf.fill([self.memory_size, self.val_depth], .0), 0),
                [num_updates, 1, 1]))
    update_logits = tf.scatter_update(
        self.mean_logits, entries_to_reset,
        tf.tile(tf.expand_dims(
            tf.fill([self.memory_size], .0), 0),
                [num_updates, 1]))
    reset_op = tf.group([update_vals, update_logits])
    return reset_op 

def Uniform(name=None):
    X = tf.placeholder(config.dtype, name=name)

    Distribution.logp = tf.fill(tf.shape(X), config.dtype(0))

    def integral(lower, upper):
        return tf.cond(
            tf.logical_or(
                tf.is_inf(tf.cast(lower, config.dtype)),
                tf.is_inf(tf.cast(upper, config.dtype))
            ),
            lambda: tf.constant(1, dtype=config.dtype),
            lambda: tf.cast(upper, config.dtype) - tf.cast(lower, config.dtype),
        )

    Distribution.integral = integral

    return X 

def _build(self, obs_input, name=None):
        """Build model given input placeholder(s).

        Args:
            obs_input (tf.Tensor): Tensor input for state.
            name (str): Inner model name, also the variable scope of the
                inner model, if exist. One example is
                garage.tf.models.Sequential.

        Return:
            tf.Tensor: Tensor output of the model.

        """
        del name
        height_size = obs_input.get_shape().as_list()[1]
        width_size = obs_input.get_shape().as_list()[2]
        for filter_iter, stride in zip(self.filters, self.strides):
            if self.padding == 'SAME':
                height_size = int((height_size + stride - 1) / stride)
                width_size = int((width_size + stride - 1) / stride)
            else:
                height_size = int(
                    (height_size - filter_iter[1][0]) / stride) + 1
                width_size = int((width_size - filter_iter[1][1]) / stride) + 1
        flatten_shape = height_size * width_size * self.filters[-1][0]
        return_var = tf.compat.v1.get_variable(
            'return_var', (), initializer=tf.constant_initializer(0.5))
        return tf.fill((tf.shape(obs_input)[0], flatten_shape), return_var) 

def set_logp_to_neg_inf(X, logp, bounds):
    """Set `logp` to negative infinity when `X` is outside the allowed bounds.

    # Arguments
        X: tensorflow.Tensor
            The variable to apply the bounds to
        logp: tensorflow.Tensor
            The log probability corrosponding to `X`
        bounds: list of `Region` objects
            The regions corrosponding to allowed regions of `X`

    # Returns
        logp: tensorflow.Tensor
            The newly bounded log probability
    """
    conditions = []
    for l, u in bounds:
        lower_is_neg_inf = not isinstance(l, tf.Tensor) and np.isneginf(l)
        upper_is_pos_inf = not isinstance(u, tf.Tensor) and np.isposinf(u)

        if not lower_is_neg_inf and upper_is_pos_inf:
            conditions.append(tf.greater(X, l))
        elif lower_is_neg_inf and not upper_is_pos_inf:
            conditions.append(tf.less(X, u))
        elif not (lower_is_neg_inf or upper_is_pos_inf):
            conditions.append(tf.logical_and(tf.greater(X, l), tf.less(X, u)))

    if len(conditions) > 0:
        is_inside_bounds = conditions[0]
        for condition in conditions[1:]:
            is_inside_bounds = tf.logical_or(is_inside_bounds, condition)

        logp = tf.select(
            is_inside_bounds,
            logp,
            tf.fill(tf.shape(X), config.dtype(-np.inf))
        )

    return logp 

def createColorfulTestImage(self):
    ch255 = tf.fill([1, 100, 200, 1], tf.constant(255, dtype=tf.uint8))
    ch128 = tf.fill([1, 100, 200, 1], tf.constant(128, dtype=tf.uint8))
    ch0 = tf.fill([1, 100, 200, 1], tf.constant(0, dtype=tf.uint8))
    imr = tf.concat([ch255, ch0, ch0], 3)
    img = tf.concat([ch255, ch255, ch0], 3)
    imb = tf.concat([ch255, ch0, ch255], 3)
    imw = tf.concat([ch128, ch128, ch128], 3)
    imu = tf.concat([imr, img], 2)
    imd = tf.concat([imb, imw], 2)
    im = tf.concat([imu, imd], 1)
    return im 

def version_9(cls, node, **kwargs):
    tensor_dict = kwargs["tensor_dict"]
    input_tensor = tensor_dict[node.inputs[0]]
    input_shape = tf.shape(input_tensor, out_type=tf.int64)

    # handle defaults for attributes
    lambd = node.attrs["lambd"] if "lambd" in node.attrs else 0.5
    bias = node.attrs["bias"] if "bias" in node.attrs else 0.0

    # make tensors in the right shape
    lambd_tensor = tf.fill(input_shape, tf.constant(lambd, input_tensor.dtype))
    lambd_neg_tensor = tf.fill(input_shape,
                               tf.constant(lambd * -1, input_tensor.dtype))
    bias_tensor = tf.fill(input_shape, tf.constant(bias, input_tensor.dtype))
    zeros_tensor = tf.zeros(input_shape, input_tensor.dtype)

    # prepare return values and conditions
    input_plus = tf.add(input_tensor, bias_tensor)
    input_minus = tf.subtract(input_tensor, bias_tensor)
    greater_cond = tf.greater(input_tensor, lambd_tensor)
    less_cond = tf.less(input_tensor, lambd_neg_tensor)

    return [
        tf.where(less_cond, input_plus,
                 tf.where(greater_cond, input_minus, zeros_tensor))
    ] 

def _graph_fn_get(self, time_percentage=None):
        if get_backend() == "tf":
            if time_percentage is not None:
                return tf.fill(tf.shape(time_percentage), self.from_)
            else:
                return self.from_
        elif get_backend() == "pytorch":
            return torch.full(time_percentage.size(), self.from_)