def batch_parse_tf_example(batch_size, layout, example_batch, use_bf16 = True):
"""
Args:
batch_size: batch size
layout: 'nchw' or 'nhwc'
example_batch: a batch of tf.Example
Returns:
A tuple (feature_tensor, dict of output tensors)
"""
planes = dual_net.get_features_planes()
features = {
'x': tf.compat.v1.FixedLenFeature([], tf.string),
'pi': tf.compat.v1.FixedLenFeature([], tf.string),
'outcome': tf.compat.v1.FixedLenFeature([], tf.float32),
}
parsed = tf.compat.v1.parse_example(example_batch, features)
x = tf.compat.v1.decode_raw(parsed['x'], tf.uint8)
if use_bf16:
x = tf.cast(x, tf.bfloat16)
else:
x = tf.cast(x, tf.float32)
if layout == 'nhwc':
shape = [batch_size, go.N, go.N, planes]
else:
shape = [batch_size, planes, go.N, go.N]
x = tf.reshape(x, shape)
pi = tf.compat.v1.decode_raw(parsed['pi'], tf.float32)
pi = tf.reshape(pi, [batch_size, go.N * go.N + 1])
outcome = parsed['outcome']
outcome.set_shape([batch_size])
return x, {'pi_tensor': pi, 'value_tensor': outcome}
def create_random_data(self, num_examples):
planes = dual_net.get_features_planes()
raw_data = []
for _ in range(num_examples):
feature = (256 * np.random.random([
go.N, go.N, planes])).astype(np.uint8)
pi = np.random.random([go.N * go.N + 1]).astype(np.float32)
value = np.random.random()
raw_data.append((feature, pi, value))
return raw_data
def batch_parse_tf_example(batch_size, example_batch):
"""
Args:
example_batch: a batch of tf.Example
Returns:
A tuple (feature_tensor, dict of output tensors)
"""
planes = dual_net.get_features_planes()
features = {
'x': tf.FixedLenFeature([], tf.string),
'pi': tf.FixedLenFeature([], tf.string),
'outcome': tf.FixedLenFeature([], tf.float32),
}
parsed = tf.parse_example(example_batch, features)
x = tf.decode_raw(parsed['x'], tf.uint8)
x = tf.cast(x, tf.float32)
x = tf.reshape(x, [batch_size, go.N, go.N, planes])
pi = tf.decode_raw(parsed['pi'], tf.float32)
pi = tf.reshape(pi, [batch_size, go.N * go.N + 1])
outcome = parsed['outcome']
outcome.set_shape([batch_size])
return x, {'pi_tensor': pi, 'value_tensor': outcome}