def run_typing_test(topics):
"""Measure typing speed and accuracy on the command line."""
paragraphs = lines_from_file('data/sample_paragraphs.txt')
select = lambda p: True
if topics:
select = about(topics)
i = 0
while True:
reference = choose(paragraphs, select, i)
if not reference:
print('No more paragraphs about', topics, 'are available.')
return
print('Type the following paragraph and then press enter/return.')
print(
'If you only type part of it, you will be scored only on that part.\n'
)
print(reference)
print()
start = datetime.now()
typed = input()
if not typed:
print('Goodbye.')
return
print()
elapsed = (datetime.now() - start).total_seconds()
print("Nice work!")
print('Words per minute:', wpm(typed, elapsed))
print('Accuracy: ', accuracy(typed, reference))
print('\nPress enter/return for the next paragraph or type q to quit.')
if input().strip() == 'q':
return
i += 1
# Training
image_size = 112
batch_size = 32
num_epochs = 20
epoch_size = 28747
train_enqueue_steps = 50
save_steps = 200 # Number of steps to perform saving checkpoints
test_steps = 20 # Number of times to test for test accuracy
start_test_step = 50
max_checkpoints_to_keep = 2
save_dir = "/home/aiteam/quan/checkpoints/ucf101"
train_data_reader = lines_from_file(train_txt, repeat=True)
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, num_frames),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int64, shape=(None, ), name='labels')
train_input_queue = data_flow_ops.FIFOQueue(capacity=10000,
dtypes=[tf.string, tf.int64],
shapes=[(num_frames, ), ()])
train_enqueue_op = train_input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder])
frames_batch, labels_batch = input_pipeline(train_input_queue,
batch_size=batch_size,
return resized_image
def visualize_frame_batches(frame_batches, num_videos, num_frames):
for i in range(num_videos):
print("Visualize video ", i)
video = frame_batches[i, :, :, :, :]
for j in range(num_frames):
frame = video[j, :, :, :]
frame = cv2.cvtColor(frame, cv2.COLOR_RGB2BGR)
cv2.imshow("frame", frame)
cv2.waitKey(100)
if __name__ == "__main__":
num_frames = 10
root_folder = "/home/ubuntu/datasets/ucf101/train/"
data_reader = lines_from_file("/home/ubuntu/datasets/ucf101/train.txt", repeat=True)
# image_paths, labels = sample_videos(data_reader, root_folder=root_folder,
# num_samples=3, num_frames=num_frames)
#
# for i in range(3):
# print(image_paths[i], labels[i])
image_paths_placeholder = tf.placeholder(tf.string, shape=(None, num_frames), name='image_paths')
labels_placeholder = tf.placeholder(tf.int64, shape=(None, ), name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=10000,
dtypes=[tf.string, tf.int64],
shapes=[(num_frames,), ()])
enqueue_op = input_queue.enqueue_many([image_paths_placeholder, labels_placeholder])
