I am constructing a neural network where the output I intend to predict simply has the following conditions implemented using tf.cond:
def net(self, x, y):
v1_out = self.neural_net(tf.concat([x,y], 1), self.weights, self.biases)
v1 = v1_out[:,0:1]
S_n = tf.exp(-(x*x)/2.)
S_n = tf.cond(x[0][0] < 0.2, lambda: 1.0, lambda: S_n)
S_n = tf.cond(S_n[0][0] < 0.01, lambda: 0.01, lambda: S_n)
S_n = tf.cond(v1[0][0] > 5.0, lambda: 0.0, lambda: S_n)
v1 = S_n*v1
return v1
Everything appears to initialize and begin training well. But then in the midst of training after a few hundred iterations, the following error appears due to the above tf.cond statements. Are there any thoughts on why specifically this is arising and how to fix this error?
2020-06-06 09:59:31.745901: W tensorflow/core/framework/op_kernel.cc:1192] Invalid argument: Index out of range using input dim 0; input has only 0 dims
...
tensorflow.python.framework.errors_impl.InvalidArgumentError: Index out of range using input dim 0; input has only 0 dims
[[Node: strided_slice_3 = StridedSlice[Index=DT_INT32, T=DT_FLOAT, begin_mask=0, ellipsis_mask=0, end_mask=0, new_axis_mask=0, shrink_axis_mask=1, _device="/job:localhost/replica:0/task:0/device:CPU:0"](cond/Merge, strided_slice_1/stack, strided_slice_1/stack_1, strided_slice_1/stack_1)]]
...
Caused by op 'strided_slice_3', defined at:
...
File "/home/software/sloan/local/lib/tensorflow/cpu/py36/1.4.1/tensorflow/python/framework/ops.py", line 1470, in __init__
self._traceback = self._graph._extract_stack() # pylint: disable=protected-access
InvalidArgumentError (see above for traceback): Index out of range using input dim 0; input has only 0 dims
[[Node: strided_slice_3 = StridedSlice[Index=DT_INT32, T=DT_FLOAT, begin_mask=0, ellipsis_mask=0, end_mask=0, new_axis_mask=0, shrink_axis_mask=1, _device="/job:localhost/replica:0/task:0/device:CPU:0"](cond/Merge, strided_slice_1/stack, strided_slice_1/stack_1, strided_slice_1/stack_1)]]
Minimal example code capable of replicating the problem:
import numpy as np
import tensorflow as tf
end_it = 10000 #number of iterations
frac_train = 1.0 #randomly sampled fraction of data to create training set
frac_sample_train = 0.1 #randomly sampled fraction of data from training set to train in batches
layers = [2, 20, 20, 20, 20, 20, 20, 20, 20, 1]
len_data = 10000
x_x = np.array([np.linspace(0.,1.,len_data)])
x_y = np.array([np.linspace(0.,1.,len_data)])
y_true = np.array([np.linspace(-1.,1.,len_data)])
N_train = int(frac_train*len_data)
idx = np.random.choice(len_data, N_train, replace=False)
x_train = x_x.T[idx,:]
y_train = x_y.T[idx,:]
v1_train = y_true.T[idx,:]
sample_batch_size = int(frac_sample_train*N_train)
np.random.seed(1234)
tf.set_random_seed(1234)
import logging
logging.getLogger('tensorflow').setLevel(logging.ERROR)
tf.logging.set_verbosity(tf.logging.ERROR)
class NeuralNet:
def __init__(self, x, y, v1, layers):
X = np.concatenate([x, y], 1)
self.lb = X.min(0)
self.ub = X.max(0)
self.X = X
self.x = X[:,0:1]
self.y = X[:,1:2]
self.v1 = v1
self.layers = layers
self.weights, self.biases = self.initialize_NN(layers)
self.sess = tf.Session(config=tf.ConfigProto(allow_soft_placement=False,
log_device_placement=False))
self.x_tf = tf.placeholder(tf.float32, shape=[None, self.x.shape[1]])
self.y_tf = tf.placeholder(tf.float32, shape=[None, self.y.shape[1]])
self.v1_tf = tf.placeholder(tf.float32, shape=[None, self.v1.shape[1]])
self.v1_pred = self.net(self.x_tf, self.y_tf)
self.loss = tf.reduce_mean(tf.square(self.v1_tf - self.v1_pred))
self.optimizer = tf.contrib.opt.ScipyOptimizerInterface(self.loss,
method = 'L-BFGS-B',
options = {'maxiter': 50,
'maxfun': 50000,
'maxcor': 50,
'maxls': 50,
'ftol' : 1.0 * np.finfo(float).eps})
init = tf.global_variables_initializer()
self.sess.run(init)
def initialize_NN(self, layers):
weights = []
biases = []
num_layers = len(layers)
for l in range(0,num_layers-1):
W = self.xavier_init(size=[layers[l], layers[l+1]])
b = tf.Variable(tf.zeros([1,layers[l+1]], dtype=tf.float32), dtype=tf.float32)
weights.append(W)
biases.append(b)
return weights, biases
def xavier_init(self, size):
in_dim = size[0]
out_dim = size[1]
xavier_stddev = np.sqrt(2/(in_dim + out_dim))
return tf.Variable(tf.truncated_normal([in_dim, out_dim], stddev=xavier_stddev), dtype=tf.float32)
def neural_net(self, X, weights, biases):
num_layers = len(weights) + 1
H = 2.0*(X - self.lb)/(self.ub - self.lb) - 1.0
for l in range(0,num_layers-2):
W = weights[l]
b = biases[l]
H = tf.tanh(tf.add(tf.matmul(H, W), b))
W = weights[-1]
b = biases[-1]
Y = tf.add(tf.matmul(H, W), b)
return Y
def net(self, x, y):
v1_out = self.neural_net(tf.concat([x,y], 1), self.weights, self.biases)
v1 = v1_out[:,0:1]
S_n = tf.exp(-(x*x)/2.)
S_n = tf.cond(x[0][0] < 0.2, lambda: 1.0, lambda: S_n)
S_n = tf.cond(S_n[0][0] < 0.01, lambda: 0.01, lambda: S_n)
S_n = tf.cond(v1[0][0] > 5.0, lambda: 0.0, lambda: S_n)
v1 = S_n*v1
return v1
def callback(self, loss):
global Nfeval
print(str(Nfeval)+' - Loss in loop: %.3e' % (loss))
Nfeval += 1
def fetch_minibatch(self, x_in, y_in, den_in, N_train_sample):
idx_batch = np.random.choice(len(x_in), N_train_sample, replace=False)
x_batch = x_in[idx_batch,:]
y_batch = y_in[idx_batch,:]
v1_batch = den_in[idx_batch,:]
return x_batch, y_batch, v1_batch
def train(self, end_it):
it = 0
while it < end_it:
x_res_batch, y_res_batch, v1_res_batch = self.fetch_minibatch(self.x, self.y, self.v1, sample_batch_size) # Fetch residual mini-batch
tf_dict = {self.x_tf: x_res_batch, self.y_tf: y_res_batch,
self.v1_tf: v1_res_batch}
self.optimizer.minimize(self.sess,
feed_dict = tf_dict,
fetches = [self.loss],
loss_callback = self.callback)
def predict(self, x_star, y_star):
tf_dict = {self.x_tf: x_star, self.y_tf: y_star}
v1_star = self.sess.run(self.v1_pred, tf_dict)
return v1_star
model = NeuralNet(x_train, y_train, v1_train, layers)
Nfeval = 1
model.train(end_it)
Aucun commentaire:
Enregistrer un commentaire