jeakwon

from sqlalchemy import create_engine

db = create_engine('postgresql://user:password@localhost:port/database')
for i in range(1,2000):
    conn = db.connect()
    #some simple data operations
    conn.close()
db.dispose()

jeakwon

import time
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

from torch.utils.data import DataLoader, random_split
from torchvision import datasets, transforms

from activation import load_actF

jeakwon

# Example:
# $ python argparse_dict_argument.py --env a=b --env aa=bb
# Namespace(env={'a': 'b', 'aa': 'bb'})

import argparse

parser = argparse.ArgumentParser()
parser.add_argument('--env', action = type('', (argparse.Action, ), dict(__call__ = lambda a, p, n, v, o: getattr(n, a.dest).update(dict([v.split('=')])))), default = {}) # anonymously subclassing argparse.Action
print parser.parse_args()

jeakwon

import time
import torch

class Tag:
    def __init__(self, topk=(1,), verbose=True):
        self.__topk = topk
        self.__verbose = verbose

    def __call__(self, data_loader):
        self.__time_start = time.time()

jeakwon

import time
import torch

class AverageMeter:
    def __init__(self):
        self.reset()

    def reset(self):
        self.val = 0
        self.avg = 0

jeakwon

"""
Create train, valid, test iterators for CIFAR-10 [1].
Easily extended to MNIST, CIFAR-100 and Imagenet.
[1]: https://discuss.pytorch.org/t/feedback-on-pytorch-for-kaggle-competitions/2252/4
"""

import torch
import numpy as np

from utils import plot_images

jeakwon

# Cutting .wav file from start to end
# ffmpeg -i <in_file> -ss <start_sec> -to <end_sec> -c copy <out_file>
# ffmpeg -i 1.mp3 -ss 0 -to 60 -c copy 2.mp3 (example of use)

# Generating mono .wav file
# ffmpeg -i <in_file> -acodec pcm_s16le -ac 1 -ar 16000 <out_file>
# ffmpeg -i 2.mp3 -acodec pcm_s16le -ac 1 -ar 16000 2.wav

# Combine both
# ffmpeg -i 1.mp3 -ss 0 -to 60 -c -acodec pcm_s16le -ac 1 -ar 16000 copy out.wav

jeakwon

import os
os.environ["GOOGLE_APPLICATION_CREDENTIALS"]="<credential_file>.json"

def transcribe_gcs_with_word_time_offsets(gcs_uri):
    """Transcribe the given audio file asynchronously and output the word time
    offsets."""
    from google.cloud import speech
    from google.cloud.speech import enums
    from google.cloud.speech import types
    client = speech.SpeechClient()

jeakwon

def runtime(func):
    @wraps(func)
    def decorated(*args, **kwargs):
        # Decorate front of func
        before  = time.time()

        output  = func(*args, **kwargs)

        # Decorate back of func
        after   = time.time()

jeakwon

# -*- coding: utf-8 -*-
import logging.handlers

class Setting:
    """로거 세팅 클래스

        ::

            Setting.LEVEL = logging.INFO # INFO 이상만 로그를 작성
    """