def precision_recall(data,
cls,
prob_threshold,
iou_treshold,
pred_field='predictions',
gt_field='ground_truth'):
TP = 0.0
TPFP = 0.0 # total positive results / pred
TPFN = 0.0 # total existing cases / rel
for obj in data:
ground_truth = (obj[gt_field]
| mp.where(lambda x: x['tag'] == cls)
| mp.as_list)
TPFN += len(ground_truth)
predictions = (
obj[pred_field]
|
mp.where(lambda x: x['tag'] == cls and x['prob'] > prob_threshold)
| mp.as_list)
for gt_box in ground_truth:
pred_boxes = (
predictions
| mp.apply(
['x1', 'y1', 'width', 'height'], 'iou',
lambda x: intersection_over_union(x, (gt_box['x1'], gt_box[
'y1'], gt_box['width'], gt_box['height'])))
| mp.filter('iou', lambda x: x < iou_treshold)
| mp.as_list)
if len(pred_boxes) > 0:
TP += 1
TPFP += len(pred_boxes)
return (float(TP > 0) if TPFP == 0 else TP / TPFP,
float(TP > 0) if TPFN == 0 else TP / TPFN)
def coco_to_custom_vision(stream, project_id, trainer, data_dir):
stream = stream | mp.as_list
tags = stream | mp.select_field('class_id') | mp.dedup() | mp.as_list
cv_tags = {tag: trainer.create_tag(project_id, tag) for tag in tags}
stream = (
stream
| mp.apply(['width', 'height', 'ground_truth'], 'ground_truth',
lambda x: x[2]
| mp.where(lambda box: (box['width'] >= x[0] * 0.1) and
(box['height'] >= x[1] * 0.1))
| mp.as_list)
| mp.filter('ground_truth', lambda x: len(x) > 0)
| mp.apply(
['width', 'height', 'ground_truth'], 'regions', lambda x: x[2]
| mp.select(lambda box: Region(tag_id=cv_tags[box['tag']].id,
left=box['x1'] / x[0],
top=box['y1'] / x[1],
width=box['width'] / x[0],
height=box['height'] / x[1]))
| mp.as_list)
| mp.apply(['filename', 'regions'], 'tagged_img',
lambda x: ImageFileCreateEntry(
name=x[0],
contents=open(join(data_dir, x[0]), mode="rb").read(),
regions=x[1]))
| mp.as_list)
tagged_images_with_regions = stream | mp.select_field(
'tagged_img') | mp.as_list
for i in range(0, len(tagged_images_with_regions), 50):
trainer.create_images_from_files(
project_id, images=tagged_images_with_regions[i:i + 50])
from pipe import Pipe
from moviepy.editor import *
import numpy as np
import itertools
import cv2
import math
import matplotlib.pyplot as plt
import keras
from keras.models import Sequential
from keras.layers import *
from keras.regularizers import l2
# Get list of test videos from matches.json
test_names = (from_json(os.path.join(source_dir, 'matches.json'))
| mp.where(lambda x: 'Test' in x.keys() and int(x['Test']) > 0)
| mp.apply(['Id', 'Half'], 'pattern',
lambda x: "{}_{}_".format(x[0], x[1]))
| mp.select_field('pattern')
| mp.as_list)
data = (mp.get_datastream(data_dir, ext=".resized.mp4")
| datasplit_by_pattern(test_pattern=test_names)
| stratify_sample_tt(shuffle=True)
| summary()
| mp.take(1000)
| mp.iter('filename', lambda x: print("Processing {}".format(x)))
| mp.apply(
'filename', 'aud',
lambda x: np.load(x.replace('.resized.mp4', '.audiofeatures.npy')))
| normalize_npy_value('aud', interval=(-1, 1))
parser.add_argument("dir",
help="Directory of pictures to be processed",
default=".")
parser.add_argument("--facegroup",
help="Name of face group to use for face recognition",
default="")
args = parser.parse_args()
face.BaseUrl.set(conf['FaceApi']['Endpoint'])
face.Key.set(conf['FaceApi']['Key'])
print("Face API Calling Utility")
files = mp.get_files(
args.dir) | mp.where(lambda x: not x.endswith('.json')) | mp.as_list
for x in files:
print(" + {}".format(x), end='')
jsfn = os.path.splitext(x)[0] + '.json'
if os.path.isfile(jsfn):
print(" -> skip")
continue
res = []
try:
res = face.face.detect(
x, True, True,
'age,gender,headPose,smile,facialHair,glasses,emotion,hair,makeup,occlusion,accessories,blur'
)
except:
pass