Python tensorflow.neg() Examples

def testSideEffect(self):
    a = tf.constant(1)
    b = tf.constant(1)
    c = tf.add(a, b)
    with tf.control_dependencies([c]):
      d = tf.constant(42)
    n = tf.neg(c)

    shared = []

    def sub(t):
      shared.append(t)
      return t

    c = subscribe.subscribe(c, lambda t: tf.py_func(sub, [t], [t.dtype]))

    with self.test_session() as sess:
      c_out = sess.run([c])
      n_out = sess.run([n])
      d_out = sess.run([d])

    self.assertEquals(n_out, [-2])
    self.assertEquals(c_out, [2])
    self.assertEquals(d_out, [42])
    self.assertEquals(shared, [2, 2, 2]) 

def init_ops_for_training(self, critic):
    # actors gradients are the gradients for it's output w.r.t it's vars using initial
    # gradients provided by critic. this requires that critic was init'd with an
    # input_action = actor.output_action (which is natural anyway)
    # we wrap the optimiser in namespace since we don't want this as part of copy to
    # target networks.
    # note that we negate the gradients from critic since we are trying to maximise
    # the q values (not minimise like a loss)
    with tf.variable_scope("optimiser"):
      gradients = tf.gradients(self.output_action,
                               self.trainable_model_vars(),
                               tf.neg(critic.q_gradients_wrt_actions()))
      gradients = zip(gradients, self.trainable_model_vars())
      # potentially clip and wrap with debugging
      gradients = util.clip_and_debug_gradients(gradients, opts)
      # apply
      optimiser = tf.train.GradientDescentOptimizer(opts.actor_learning_rate)
      self.train_op = optimiser.apply_gradients(gradients) 

def testFloatBasic(self):
    x = np.arange(-3, 3).reshape(1, 3, 2).astype(np.float32)
    y = (x + .5).astype(np.float32)     # no zero
    z = (x + 15.5).astype(np.float32)   # all positive
    k = np.arange(-0.90, 0.90, 0.25).astype(np.float32) # between -1 and 1

    self._compareBoth(x, np.abs, tf.abs)
    self._compareBoth(x, np.abs, _ABS)
    self._compareBoth(x, np.negative, tf.neg)
    self._compareBoth(x, np.negative, _NEG)
    self._compareBoth(y, self._inv, tf.inv)
    self._compareBoth(x, np.square, tf.square)
    self._compareBoth(z, np.sqrt, tf.sqrt)
    self._compareBoth(z, self._rsqrt, tf.rsqrt)
    self._compareBoth(x, np.exp, tf.exp)
    self._compareBoth(z, np.log, tf.log)
    self._compareBoth(z, np.log1p, tf.log1p)
    self._compareBoth(x, np.tanh, tf.tanh)
    self._compareBoth(x, self._sigmoid, tf.sigmoid)
    self._compareBoth(y, np.sign, tf.sign)
    self._compareBoth(x, np.sin, tf.sin)
    self._compareBoth(x, np.cos, tf.cos)
    self._compareBoth(k, np.arcsin, tf.asin)
    self._compareBoth(k, np.arccos, tf.acos)
    self._compareBoth(x, np.arctan, tf.atan)
    self._compareBoth(x, np.tan, tf.tan)
    self._compareBoth(
        y,
        np.vectorize(self._replace_domain_error_with_inf(math.lgamma)),
        tf.lgamma)
    self._compareBoth(x, np.vectorize(math.erf), tf.erf)
    self._compareBoth(x, np.vectorize(math.erfc), tf.erfc)

    self._compareBothSparse(x, np.abs, tf.abs)
    self._compareBothSparse(x, np.negative, tf.neg)
    self._compareBothSparse(x, np.square, tf.square)
    self._compareBothSparse(z, np.sqrt, tf.sqrt, tol=1e-3)
    self._compareBothSparse(x, np.tanh, tf.tanh)
    self._compareBothSparse(y, np.sign, tf.sign)
    self._compareBothSparse(x, np.vectorize(math.erf), tf.erf) 

def testFloatEmpty(self):
    x = np.empty((2, 0, 5), dtype=np.float32)
    self._compareBoth(x, np.abs, tf.abs)
    self._compareBoth(x, np.abs, _ABS)
    self._compareBoth(x, np.negative, tf.neg)
    self._compareBoth(x, np.negative, _NEG)
    self._compareBoth(x, self._inv, tf.inv)
    self._compareBoth(x, np.square, tf.square)
    self._compareBoth(x, np.sqrt, tf.sqrt)
    self._compareBoth(x, self._rsqrt, tf.rsqrt)
    self._compareBoth(x, np.exp, tf.exp)
    self._compareBoth(x, np.log, tf.log)
    self._compareBoth(x, np.log1p, tf.log1p)
    self._compareBoth(x, np.tanh, tf.tanh)
    self._compareBoth(x, self._sigmoid, tf.sigmoid)
    self._compareBoth(x, np.sign, tf.sign)
    self._compareBoth(x, np.sin, tf.sin)
    self._compareBoth(x, np.cos, tf.cos)
    # Can't use vectorize below, so just use some arbitrary function
    self._compareBoth(x, np.sign, tf.lgamma)
    self._compareBoth(x, np.sign, tf.erf)
    self._compareBoth(x, np.sign, tf.erfc)
    self._compareBoth(x, np.tan, tf.tan)
    self._compareBoth(x, np.arcsin, tf.asin)
    self._compareBoth(x, np.arccos, tf.acos)
    self._compareBoth(x, np.arctan, tf.atan)

    self._compareBothSparse(x, np.abs, tf.abs)
    self._compareBothSparse(x, np.negative, tf.neg)
    self._compareBothSparse(x, np.square, tf.square)
    self._compareBothSparse(x, np.sqrt, tf.sqrt, tol=1e-3)
    self._compareBothSparse(x, np.tanh, tf.tanh)
    self._compareBothSparse(x, np.sign, tf.sign)
    self._compareBothSparse(x, np.sign, tf.erf) 

def testDoubleBasic(self):
    x = np.arange(-3, 3).reshape(1, 3, 2).astype(np.float64)
    y = (x + .5).astype(np.float64)    # no zero
    z = (x + 15.5).astype(np.float64)  # all positive
    k = np.arange(-0.90, 0.90, 0.35).reshape(1, 3, 2).astype(np.float64) # between -1 and 1
    self._compareBoth(x, np.abs, tf.abs)
    self._compareBoth(x, np.abs, _ABS)
    self._compareBoth(x, np.negative, tf.neg)
    self._compareBoth(x, np.negative, _NEG)
    self._compareBoth(y, self._inv, tf.inv)
    self._compareBoth(x, np.square, tf.square)
    self._compareBoth(z, np.sqrt, tf.sqrt)
    self._compareBoth(z, self._rsqrt, tf.rsqrt)
    self._compareBoth(x, np.exp, tf.exp)
    self._compareBoth(z, np.log, tf.log)
    self._compareBoth(z, np.log1p, tf.log1p)
    self._compareBoth(x, np.tanh, tf.tanh)
    self._compareBoth(x, self._sigmoid, tf.sigmoid)
    self._compareBoth(y, np.sign, tf.sign)
    self._compareBoth(x, np.sin, tf.sin)
    self._compareBoth(x, np.cos, tf.cos)
    self._compareBoth(
        y,
        np.vectorize(self._replace_domain_error_with_inf(math.lgamma)),
        tf.lgamma)
    self._compareBoth(x, np.vectorize(math.erf), tf.erf)
    self._compareBoth(x, np.vectorize(math.erfc), tf.erfc)
    self._compareBoth(x, np.arctan, tf.atan)
    self._compareBoth(k, np.arcsin, tf.asin)
    self._compareBoth(k, np.arccos, tf.acos)
    self._compareBoth(k, np.tan, tf.tan)

    self._compareBothSparse(x, np.abs, tf.abs)
    self._compareBothSparse(x, np.negative, tf.neg)
    self._compareBothSparse(x, np.square, tf.square)
    self._compareBothSparse(z, np.sqrt, tf.sqrt, tol=1e-3)
    self._compareBothSparse(x, np.tanh, tf.tanh)
    self._compareBothSparse(y, np.sign, tf.sign)
    self._compareBothSparse(x, np.vectorize(math.erf), tf.erf) 

def testHalfBasic(self):
    x = np.arange(-3, 3).reshape(1, 3, 2).astype(np.float16)
    y = (x + .5).astype(np.float16)    # no zero
    z = (x + 15.5).astype(np.float16)  # all positive
    self._compareBoth(x, np.abs, tf.abs)
    self._compareBoth(x, np.abs, _ABS)
    self._compareBoth(x, np.negative, tf.neg)
    self._compareBoth(x, np.negative, _NEG)
    self._compareBoth(y, self._inv, tf.inv)
    self._compareBoth(x, np.square, tf.square)
    self._compareBoth(z, np.sqrt, tf.sqrt)
    self._compareBoth(z, self._rsqrt, tf.rsqrt)
    self._compareBoth(x, np.exp, tf.exp)
    self._compareBoth(z, np.log, tf.log)
    self._compareBoth(z, np.log1p, tf.log1p)
    self._compareBoth(x, np.tanh, tf.tanh)
    self._compareBoth(x, self._sigmoid, tf.sigmoid)
    self._compareBoth(y, np.sign, tf.sign)
    self._compareBoth(x, np.sin, tf.sin)
    self._compareBoth(x, np.cos, tf.cos)
    self._compareBoth(
        y,
        np.vectorize(self._replace_domain_error_with_inf(math.lgamma)),
        tf.lgamma)
    self._compareBoth(x, np.vectorize(math.erf), tf.erf)
    self._compareBoth(x, np.vectorize(math.erfc), tf.erfc)

    self._compareBothSparse(x, np.abs, tf.abs)
    self._compareBothSparse(x, np.negative, tf.neg)
    self._compareBothSparse(x, np.square, tf.square)
    self._compareBothSparse(z, np.sqrt, tf.sqrt, tol=1e-3)
    self._compareBothSparse(x, np.tanh, tf.tanh)
    self._compareBothSparse(y, np.sign, tf.sign)
    self._compareBothSparse(x, np.vectorize(math.erf), tf.erf, tol=1e-3) 

def testInt32Basic(self):
    x = np.arange(-6, 6, 2).reshape(1, 3, 2).astype(np.int32)
    self._compareCpu(x, np.abs, tf.abs)
    self._compareCpu(x, np.abs, _ABS)
    self._compareBoth(x, np.negative, tf.neg)
    self._compareBoth(x, np.negative, _NEG)
    self._compareBoth(x, np.square, tf.square)
    self._compareCpu(x, np.sign, tf.sign)

    self._compareBothSparse(x, np.abs, tf.abs)
    self._compareBothSparse(x, np.negative, tf.neg)
    self._compareBothSparse(x, np.square, tf.square)
    self._compareBothSparse(x, np.sign, tf.sign) 

def testComplex64Basic(self):
    x = np.complex(1, 1) * np.arange(-3, 3).reshape(1, 3, 2).astype(
        np.complex64)
    y = x + 0.5  # no zeros
    self._compareCpu(x, np.abs, tf.complex_abs)
    self._compareCpu(x, np.abs, _ABS)
    self._compareCpu(x, np.negative, tf.neg)
    self._compareCpu(x, np.negative, _NEG)
    self._compareCpu(y, self._inv, tf.inv)
    self._compareCpu(x, np.square, tf.square)
    self._compareCpu(y, np.sqrt, tf.sqrt)
    self._compareCpu(y, self._rsqrt, tf.rsqrt)
    self._compareCpu(x, np.exp, tf.exp)
    self._compareCpu(y, np.log, tf.log)
    self._compareCpu(y, np.log1p, tf.log1p)
    self._compareCpu(x, np.tanh, tf.tanh)
    self._compareCpu(x, self._sigmoid, tf.sigmoid)
    self._compareCpu(x, np.sin, tf.sin)
    self._compareCpu(x, np.cos, tf.cos)

    self._compareBothSparse(x, np.abs, tf.abs)
    self._compareBothSparse(x, np.negative, tf.neg)
    self._compareBothSparse(x, np.square, tf.square)
    self._compareBothSparse(x, np.sqrt, tf.sqrt, 1e-3)
    self._compareBothSparse(x, np.tanh, tf.tanh)

    # Numpy uses an incorrect definition of sign; use the right one instead.
    def complex_sign(x):
      return x / np.abs(x)
    self._compareCpu(y, complex_sign, tf.sign)
    self._compareBothSparse(y, complex_sign, tf.sign) 

def testComplex128Basic(self):
    x = np.complex(1, 1) * np.arange(-3, 3).reshape(1, 3, 2).astype(
        np.complex128)
    y = x + 0.5  # no zeros
    self._compareCpu(x, np.abs, tf.abs)
    self._compareCpu(x, np.abs, _ABS)
    self._compareCpu(x, np.negative, tf.neg)
    self._compareCpu(x, np.negative, _NEG)
    self._compareCpu(y, self._inv, tf.inv)
    self._compareCpu(x, np.square, tf.square)
    self._compareCpu(y, np.sqrt, tf.sqrt)
    self._compareCpu(y, self._rsqrt, tf.rsqrt)
    self._compareCpu(x, np.exp, tf.exp)
    self._compareCpu(y, np.log, tf.log)
    self._compareCpu(y, np.log1p, tf.log1p)
    self._compareCpu(x, np.tanh, tf.tanh)
    self._compareCpu(x, self._sigmoid, tf.sigmoid)
    self._compareCpu(x, np.sin, tf.sin)
    self._compareCpu(x, np.cos, tf.cos)

    self._compareBothSparse(x, np.abs, tf.abs)
    self._compareBothSparse(x, np.negative, tf.neg)
    self._compareBothSparse(x, np.square, tf.square)
    self._compareBothSparse(x, np.sqrt, tf.sqrt, 1e-3)
    self._compareBothSparse(x, np.tanh, tf.tanh)

    # Numpy uses an incorrect definition of sign; use the right one instead.
    def complex_sign(x):
      return x / np.abs(x)
    self._compareCpu(y, complex_sign, tf.sign)
    self._compareBothSparse(y, complex_sign, tf.sign) 

def activation(type, synapse):
    """Chooses the activation function to use."""
    if type == "sigmoid":
        return tf.sigmoid(synapse)
    elif type == "linear":
        return synapse
    elif type == "tanh":
        return tf.tanh(synapse)
    elif type == "radial":
        return tf.sqrt(tf.exp(tf.neg(tf.square(synapse)))) 

def gaussian_kernel(tensor_a, a_inputs, tensor_b, b_inputs, gamma):
    """Returns the Gaussian kernel matrix of two matrices of vectors
    element-wise."""
    cross = cross_matrices(tensor_a, a_inputs, tensor_b, b_inputs)

    kernel = tf.exp(tf.mul(tf.reduce_sum(tf.square(
        tf.sub(cross[0], cross[1])), reduction_indices=2),
        tf.neg(tf.constant(gamma, dtype=tf.float32))))

    return kernel 

def build_energy_op(self):
        with self.graph.as_default(), tf.device(self.energy_device):
            # [1, nbatch]
            e_x_0 = tf.neg((self.state_pl[0, :] ** 2) / (self.scale ** 2), name='E_x_0')
            # [ndims - 1, nbatch]
            e_x_k = tf.neg((self.state_pl[1:, :] ** 2) / tf.exp(self.state_pl[0, :]), name='E_x_k')
            # [nbatch]
            self.energy_op = tf.reduce_sum(tf.add(e_x_0, e_x_k), 0, name='energy_op') 

def __call__(self, x, l=1.0):
        grad_name = "FlipGradient%d" % self.num_calls
        @ops.RegisterGradient(grad_name)
        def _flip_gradients(op, grad):
            return [tf.neg(grad) * l]
        
        g = tf.get_default_graph()
        with g.gradient_override_map({"Identity": grad_name}):
            y = tf.identity(x)
            
        self.num_calls += 1
        return y 

def loss_function(self):
        pos_diff = self.anchor - self.positive
        neg_diff = self.anchor - self.negative

        pos_dist = tf.reduce_sum(tf.mul(pos_diff, pos_diff), 1)
        neg_dist = tf.reduce_sum(tf.mul(neg_diff, neg_diff), 1)

        triplet = tf.add(self.ALPHA, tf.add(pos_dist, tf.neg(neg_dist)))
        return tf.reduce_sum(tf.nn.relu(triplet))