urimerhav

import os

import dotenv
import requests

dotenv.load_dotenv()


class PiiBoundingBox(object):
    def __init__(self, x1: float, y1: float, x2: float, y2: float, page: int, pii_type: str):

urimerhav

console.log('redirector!')
window.location = "www.google.com"

urimerhav

import numpy as np

def filter_accelerations(df_acceleration, filter_length):
    filter_kernel = np.ones(filter_length) / filter_length
    z = df_acceleration['z']
    filtered_z = np.convolve(z,
                             filter_kernel,
                             'same')
    df_acceleration_filtered = df_acceleration.copy()
    df_acceleration_filtered['z'] = filtered_z

urimerhav

def identify_reps(df_acceleration):
    upward_thresh = 12.0
    z_acc = df_acceleration['z']

    # We want to the start and end location of every consecutive sequence of measurement above our threshold

    # Note: scipy.signal.find_peaks would accomplish this, but feel magical. So we do out ourselves.

    segment_start_locations = []
    segment_end_locations = []

urimerhav

import pylab as plt
# calculate time difference from session start in seconds
df_acceleration['t'] = (df_acceleration['timestamp'] - df_acceleration['timestamp'].min()) / 1000

# plot the x,y,z acceleration values on the same graph
plt.figure(figsize=(8, 8))
plt.plot(df_acceleration['t'], df_acceleration['x'], label='x')
plt.plot(df_acceleration['t'], df_acceleration['y'], label='y')
plt.plot(df_acceleration['t'], df_acceleration['z'], label='z')
plt.legend()

urimerhav

import os
import pandas as pd
import pylab as plt
from moonsense.client import Client
# get all the sessions we've recorded so far
client = Client(secret_token=os.environ['MOONSENSE_SECRET_TOKEN'])
sessions = client.list_sessions()
session_acceleration_data = []
# go over every session
for session in sessions:

urimerhav

curl --location --request POST 'https://www.proproductpix.org/api/clean_manual' \
--header 'Content-Type: multipart/form-data' \
--header 'x-api-key: YOUR_API_KEY_HERE' \
--form 'photo=@/path/to/your/photo'

urimerhav

a = 5

urimerhav

def tf_blend_images(bkg_img, obj_img, object_pixels):
    object_shape = tf.shape(obj_img)
    cropped_bkg = tf.random_crop(bkg_img, object_shape)

    composed_image = (cropped_bkg * tf.cast(tf.logical_not(object_pixels), tf.uint8) +
                      obj_img * tf.cast(object_pixels, tf.uint8))

    # random saturation etc

    return composed_image

urimerhav

  def gen_item(self):
      while True:
            bkg_image = random.sample(self.all_bkg_images, 1)[0]
            png_name = self.get_complete_png(cleaned_image=False)

          yield {'png_path': str(png_name),
                 'bkg_image': bkg_image,