def fiscTrain(csv_file, model_filename):
print "Fischer training"
print csv_file
print model_filename
global model
model = recognition.get_model_from_csv(filename=csv_file,
out_model_filename=model_filename)
print model
def identifyAndTrain(image_path, csv_file, model_filename):
print "With Training"
print image_path
print csv_file
print model_filename
global model
model = recognition.get_model_from_csv(filename=csv_file,
out_model_filename=model_filename)
print model
image = open(image_path, 'rb') #open binary file in read mode
image_read = image.read()
image_data = base64.encodestring(image_read)
#print image_data
prediction = get_prediction(image_data)
return prediction
"--port",
action="store",
dest="port",
default=5000,
help="Sets the port for this server.",
required=False)
parser.add_argument('model_filename',
nargs='?',
help="Filename of the model to use or store")
# Print Usage:
print "=== Usage ==="
parser.print_help()
# Parse the Arguments:
args = parser.parse_args()
# Uh, this is ugly...
global model
# If a DataSet is given, we want to work with it:
if args.dataset:
# Learn the new model with the dataset given:
model = recognition.get_model_from_csv(
filename=args.dataset, out_model_filename=args.model_filename)
else:
model = recognition.load_model_file(args.model_filename)
# Finally start the server:
print "=== Server Log (also in %s) ===" % (LOG_FILENAME)
app.run(host=args.host,
port=args.port,
debug=True,
use_reloader=False,
threaded=False)
print "=== Description ==="
print long_description
print "=== Usage ==="
print "Usage:", usage
print "=== Server Log (also in %s) ===" % (LOG_FILENAME)
# Parse the command line:
from optparse import OptionParser
parser = OptionParser(usage=usage)
parser.add_option("-t", "--train", action="store", type="string", dest="dataset", default=None,
help="Calculates a new model from a given CSV file. CSV format: <person>;</path/to/image/folder>.")
# Split between options and arguments
(options, args) = parser.parse_args()
# Check if a model filename was passed:
if len(args) == 0:
print "Expected a facerec model to use for recognition!"
sys.exit()
# The filename of the model:
model_filename = args[0]
# Uh, this is ugly...
global model
# If a DataSet is given, we want to work with it:
if options.dataset:
# Learn the new model with the dataset given:
model = recognition.get_model_from_csv(filename=options.dataset,out_model_filename=model_filename)
else:
model = recognition.load_model_file(model_filename)
# Finally start the server:
app.run(host="0.0.0.0", port=int("5000"), debug=True, use_reloader=False, threaded=False)
)
parser.add_argument(
"-a",
"--address",
action="store",
dest="host",
default="0.0.0.0",
help="Sets the endpoint for this server.",
required=False,
)
parser.add_argument(
"-p", "--port", action="store", dest="port", default=5000, help="Sets the port for this server.", required=False
)
parser.add_argument("model_filename", nargs="?", help="Filename of the model to use or store")
# Print Usage:
print "=== Usage ==="
parser.print_help()
# Parse the Arguments:
args = parser.parse_args()
# Uh, this is ugly...
global model
# If a DataSet is given, we want to work with it:
if args.dataset:
# Learn the new model with the dataset given:
model = recognition.get_model_from_csv(filename=args.dataset, out_model_filename=args.model_filename)
else:
model = recognition.load_model_file(args.model_filename)
# Finally start the server:
print "=== Server Log (also in %s) ===" % (LOG_FILENAME)
app.run(host=args.host, port=args.port, debug=True, use_reloader=False, threaded=False)