wassname/ray
1
0
Fork 0
mirror of https://github.com/wassname/ray.git synced 2026-07-27 11:26:41 +08:00
Code Issues Packages Projects Releases Wiki Activity

enable running example apps in cluster mode (#357)

Browse Source
This commit is contained in:
Robert Nishihara
2016-08-08 16:01:13 -07:00
committed by Philipp Moritz
parent feee1de56f
commit 13df8302e6
10 changed files with 139 additions and 52 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/hyperopt/hyperopt.py
+3 -3
View File
@@ -52,7 +52,7 @@ def cnn_setup(x, y, keep_prob, lr, stddev):
# Define a remote function that takes a set of hyperparameters as well as the
# data, consructs and trains a network, and returns the validation accuracy.
@ray.remote([dict, int, np.ndarray, np.ndarray, np.ndarray, np.ndarray], [float])
def train_cnn_and_compute_accuracy(params, epochs, train_images, train_labels, validation_images, validation_labels):
def train_cnn_and_compute_accuracy(params, steps, train_images, train_labels, validation_images, validation_labels):
# Extract the hyperparameters from the params dictionary.
learning_rate = params["learning_rate"]
batch_size = params["batch_size"]
@@ -68,7 +68,7 @@ def train_cnn_and_compute_accuracy(params, epochs, train_images, train_labels, v
with tf.Session() as sess:
# Initialize the network weights.
sess.run(tf.initialize_all_variables())
for i in range(1, epochs):
for i in range(1, steps + 1):
# Fetch the next batch of data.
image_batch = get_batch(train_images, i, batch_size)
label_batch = get_batch(train_labels, i, batch_size)
@@ -82,7 +82,7 @@ def train_cnn_and_compute_accuracy(params, epochs, train_images, train_labels, v
if train_ac < 0.25:
# Compute the validation accuracy and return.
totalacc = accuracy.eval(feed_dict={x: validation_images, y: validation_labels, keep_prob: 1.0})
return totalacc
return float(totalacc)
# Training is done, compute the validation accuracy and return.
totalacc = accuracy.eval(feed_dict={x: validation_images, y: validation_labels, keep_prob: 1.0})
return float(totalacc)