def process_queue(source_directory, destination_directory, queue_directory, encoding_profile):
queue = listdir(queue_directory)
for queued_name in queue:
source_file = join(source_directory, queued_name)
if exists(source_file):
destination_file = generate_output_name(source_file, destination_directory)
generate_output(source_file, destination_file, encoding_profile)
queued_file = join(queue_directory, queued_name)
remove(queued_file)
else:
print("Warning: {0} does not exist".format(source_file))
def mirror_videos(source_directory, destination_directory, log_directory, exclusions, only, encoding_profile):
source_pattern = join(source_directory, MKV_SEARCH)
source_glob = glob(source_pattern)
source_set = set(source_glob)
destination_pattern = join(destination_directory, MP4_SEARCH)
destination_glob = glob(destination_pattern)
log_pattern = join(log_directory, MP4_SEARCH)
log_glob = glob(log_pattern)
encode_list = []
if exclusions and only:
print('mirror_videos was passed both exclusions and only')
exit_with_error()
return
if exclusions:
for exclusion in exclusions:
exclusion_pattern = join(source_directory, exclusion)
exclusion_glob = glob(exclusion_pattern)
exclusion_set = set(exclusion_glob)
source_set -= exclusion_set
elif only:
source_set = set()
for pattern in only:
only_pattern = join(source_directory, pattern)
only_glob = glob(only_pattern)
only_set = set(only_glob)
source_set |= only_set
for source_file in source_set:
destination_file = generate_output_name(source_file, destination_directory)
log_file = generate_output_name(source_file, log_directory)
if log_file in log_glob:
# We have a log of doing this conversion. We should make sure we keep the log
# intact and not convert it again.
log_glob.remove(log_file)
elif destination_file in destination_glob:
# We have the destination file, but not the log of doing it. Create the log.
create_log_file(log_file)
else:
# We haven't converted this video yet. Queue it up
encode_job = (source_file, destination_file, log_file)
encode_list.append(encode_job)
for log_file in log_glob:
print("Deleting " + log_file)
remove(log_file)
for input_file, output_file, log_file in encode_list:
generate_output(input_file, output_file, encoding_profile)
create_log_file(log_file)
def main():
args = parse_args()
testing_suite = BenchmarkSuite(args.capsule_dir, args.parallelism)
results = testing_suite.test(args.num_samples)
df = pd.DataFrame.from_records(results,
columns=BenchmarkSuite.Result._fields)
df.sort_values(by=list(df.columns), inplace=True, ignore_index=True)
output.generate_output(output=df,
csv_path=args.output_csv,
graph_path=args.output_graph)
def on_ok(self, evt):
if not self.pan.Validate():
return
params = self.pan.get_data()
data = get_query(self.report, params)
headers = data.next()
col_types = data.next()
try:
first_row = data.next()
except StopIteration:
wx.MessageBox('Sorry, no data.', 'Report', wx.OK|wx.ICON_INFORMATION)
return
out_name = self.available_outputs[self.output_doc.GetSelection()]
generate_output(self, self.report, out_name,
headers, col_types, first_row, data)
def main():
input_file = sys.argv[1]
[endpoints, caches, videos] = parse(input_file)
#solution = CacheDistributor.distribute(
# DistributionStrategy.popular_content,
# endpoints,
# caches,
# videos
#)
solution = CacheDistributor.distribute(
DistributionStrategy.active_endpoints,
endpoints,
caches,
videos
)
result = score(endpoints, solution)
generate_output(len(caches), solution)
print(result)
def open_file_dialog(self):
filename, filter = QtGui.QFileDialog.getSaveFileName(parent=self, caption='Select output file', dir='.', filter= FILE_FILTERS )
if filename:
seq = self.get_results()
output.generate_output(filename,seq)
print('start training new model')
# training
model.train()
model, loss = train(train_dir, train_dataloader, model, optimizer, loss_fn,
num_epoches, use_gpu, quiet)
print('Training complete, final loss=', loss.data)
torch.save(model.state_dict(), model_file)
print('Model saved')
# generate output
if make_validation_output:
# empty folder
for the_file in os.listdir(validation_dir):
file_path = os.path.join(validation_dir, the_file)
try:
if os.path.isfile(file_path):
os.unlink(file_path)
except Exception as e:
print(e)
generate_output(validation_dir, validation_dataloader, model, use_gpu)
print('Validation output written')
# evaluate validation outputs
mae = evaluate(validation_dir + '/normal', validation_dir + '/gt',
validation_dir + '/mask')
print('MAE = ', mae)
else:
generate_output(test_dir, test_dataloader, model, use_gpu)
print('Test output written')
