def predict_cam():
model_path = './'
global_model = load_model.ImagenetModel(model_path + 'synset.txt', model_path + 'Inception-BN')
while (True):
predictions = global_model.predict_from_cam()
print(predictions)
"""
import socket
import sys, os
import json
import logging
from timeit import default_timer as timer
import datetime
import load_model
import cv2
import csv
logging.basicConfig(
format='%(asctime)s %(name)-20s %(levelname)-5s %(message)s')
model_path = '/greengrass-machine-learning/mxnet/squeezenet/'
global_model = load_model.ImagenetModel(model_path + 'synset.txt',
model_path + 'squeezenet_v1.1')
STATS_DIRECTORY = '/home/pi/AWS/mountedStatistics'
# write local stats in a csv file
def write_local_stats(filename, stats_list):
global STATS_DIRECTORY
try:
filepath = STATS_DIRECTORY.rstrip(os.sep) + os.sep + filename
with open(filepath, 'w') as file:
writer = csv.writer(file, delimiter=',')
writer.writerows(stats_list)
except:
e = sys.exc_info()[0]
print("Exception occured during writting Statistics File: %s" % e)
#sys.exit(0)
# long-lived it will run forever when deployed to a Greengrass core. The handler
# will NOT be invoked in our example since we are executing an infinite loop.
import sys
import time
import greengrasssdk
import platform
import os
from threading import Timer
import load_model
client = greengrasssdk.client('iot-data')
#model_path = '/greengrass-machine-learning/mxnet/squeezenet/'
model_path = './mxnet_models/squeezenetv1.1/'
global_model = load_model.ImagenetModel(model_path, 'MXNET', 'synset.txt',
'squeezenet_v1.1')
#model_path = '/greengrass-machine-learning/tf/mobilenet/'
#model_path = './tf_models/mobilenetv1/'
#global_model = load_model.ImagenetModel(model_path, 'TF', 'labels.txt', 'graph.pb', 'MobilenetV1/Predictions/Reshape_1', 'CPU', [('input', (128, 128, 224, 224))])
# When deployed to a Greengrass core, this code will be executed immediately
# as a long-lived lambda function. The code will enter the infinite while loop
# below.
def greengrass_object_classification_run():
if global_model is not None:
try:
predictions = global_model.predict_from_cam()
print predictions
#publish predictions
import platform
import os
from threading import Timer
import load_model
client = greengrasssdk.client('iot-data')
#model_path = '/greengrass-machine-learning/mxnet/squeezenet/'
#model_path = './mxnet_models/squeezenetv1.1/'
#global_model = load_model.ImagenetModel(model_path, 'MXNET', 'synset.txt', 'squeezenet_v1.1')
#model_path = '/greengrass-machine-learning/tf/mobilenet/'
model_path = './tf_models/mobilenetv1/'
global_model = load_model.ImagenetModel(model_path, 'TF', 'labels.txt',
'graph.pb',
'MobilenetV1/Predictions/Reshape_1',
'CPU',
[('input', (128, 128, 224, 224))])
# When deployed to a Greengrass core, this code will be executed immediately
# as a long-lived lambda function. The code will enter the infinite while loop
# below.
def greengrass_object_classification_run():
if global_model is not None:
try:
predictions = global_model.predict_from_cam()
print predictions
#publish predictions
client.publish(topic='hello/world',
payload='New Prediction: {}'.format(
logging.basicConfig(
format='%(asctime)s %(name)-20s %(levelname)-5s %(message)s')
#model_path = '/greengrass-machine-learning/mxnet/squeezenet/'
# =============================================================================
# model_path = '.{}mxnet_models{}squeezenetv1.1{}'.format(os.sep, os.sep, os.sep)
# global_model = load_model.ImagenetModel(model_path + 'synset.txt', model_path + 'squeezenet_v1.1')
# =============================================================================
# =============================================================================
# model_path = '.{}mxnet_models{}caffenet{}'.format(os.sep, os.sep, os.sep)
# global_model = load_model.ImagenetModel(model_path + 'synset.txt', model_path + 'caffenet')
# =============================================================================
model_path = '.{}mxnet_models{}vgg16{}'.format(os.sep, os.sep, os.sep)
global_model = load_model.ImagenetModel(model_path + 'synset.txt',
model_path + 'vgg16')
STATS_DIRECTORY = './'
image_folderPath = '/home/anirban/Pictures/First2001'
# write local stats in a csv file
def write_local_stats(filename, stats_list):
global STATS_DIRECTORY
try:
filepath = STATS_DIRECTORY.rstrip(os.sep) + os.sep + filename
with open(filepath, 'w') as file:
writer = csv.writer(file, delimiter=',')
writer.writerows(stats_list)
except:
e = sys.exc_info()[0]
# will NOT be invoked in our example since the we are executing an infinite loop.
#
# This can be found on the AWS IoT Console.
import greengrasssdk
import platform
from threading import Timer
import time
import load_model
import sys
# Creating a greengrass core sdk client
client = greengrasssdk.client('iot-data')
model_path = '/greengrass-machine-learning/mxnet/inception_bn/'
global_model = load_model.ImagenetModel(model_path + 'synset.txt',
model_path + 'Inception-BN')
def greengrass_long_run():
if global_model is not None:
try:
predictions = global_model.predict_from_cam()
print predictions
#publish predictions
client.publish(topic='hello/world',
payload='New Prediction: {}'.format(
str(predictions)))
except:
e = sys.exc_info()[0]
print("Exception occured during prediction: %s" % e)
import cameratest
import load_model
print('inside main module')
print('main module name' + __name__)
#cameratest.TakePicture()
#cameratest.UploadToS3()
#source_face, matches = cameratest.compare_faces(cameratest.BUCKET, cameratest.KEY_SOURCE, cameratest.BUCKET, cameratest.KEY_TARGET)
#print("Source Face ({Confidence}%)".format(**source_face))
#for match in matches:
# print ("Target Face ({Confidence}%)".format(**match['Face']))
# print (" Similarity : {}%".format(match['Similarity']))
objtest = load_model.ImagenetModel('synset.txt', 'squeezenet_v1.1')
objtest.predict_from_file('cat.jpg')