smellslikeml

model:
  arch: video_llama
  model_type: pretrain_vicuna
  freeze_vit: True
  freeze_qformer: True
  max_txt_len: 160
  end_sym: "###"
  low_resource: True

  frozen_llama_proj: False

smellslikeml

import os
import time

import cv2
import hashlib
import numpy as np
from PIL import Image
from absl import logging
import tritonclient.http

smellslikeml

#!/usr/bin/env python3

import sys

import rospy
import rostopic

import numpy as np

import message_filters

smellslikeml

import cv2
import numpy as np
from collections import deque

# https://github.com/elliottzheng/face-detection
from face_detection import RetinaFace

# https://pyscenedetect.readthedocs.io/projects/Manual/en/latest/api/scene_manager.html#scenemanager-example
from scenedetect import VideoManager
from scenedetect import SceneManager

smellslikeml

# http://index-of.es/Varios-2/Practical%20Python%20AI%20Projects%20Mathemathical%20Models%20of%20Optimization%20Problems%20with%20Google%20OR-Tools.pdf
# coding: utf-8
"""
Example code for solving the transshipment problem. (pg. 111)

Input is an NxN numpy matrix (referenced here as D)
where row N is the demand and column N is the supply for each location 0 through N-1.

To solve, call the solve_model() function with matrix D as the input.
"""

smellslikeml

#!/usr/bin/python3
from __future__ import absolute_import, print_function
import cv2
import numpy as np
from time import sleep, time
from collections import deque

from tensorflow.keras.models import Model
from tensorflow.keras.optimizers import RMSprop
from tensorflow.keras.layers import Input, Conv2D, MaxPooling2D, UpSampling2D, ZeroPadding2D

smellslikeml

import pytube
  
def yt_download(video_id, video_path):
    video_url = 'https://youtu.be/{}'.format(video_id)
    youtube = pytube.YouTube(video_url)
    video = youtube.streams.first()
    video.download(video_path)
    print('youtube video {} saved as: \n\n {}{}.mp4'.format(video_id, video_path, video.title))

smellslikeml

from __future__ import print_function
import os
import io
import boto3
import base64
import numpy as np
import zipfile
from PIL import Image
from tflite_runtime.interpreter import Interpreter

smellslikeml

smellslikeml

#!/usr/bin/env python
import os
import sys
import requests
from bs4 import BeautifulSoup as bs
from urllib.parse import urlencode
from pytube import YouTube

qstring = sys.argv[1]
out_dir = sys.argv[2]