pzelasko

"""
Make sure to first run:
$ pip install click pandas lhotse kaldialign
"""
import click
import pandas as pd
from lhotse.serialization import load_jsonl
from kaldialign import align, bootstrap_wer_ci

EPS = '*'

pzelasko

# Note: after downloading ~250 citations further connections might be blocked by Google Scholar.

import time
import shelve

# Make sure to run: pip install scholarly tqdm
from scholarly import scholarly
from tqdm.auto import tqdm

search_query = scholarly.search_author('Piotr Żelasko')  # replace the author

pzelasko

#!/usr/bin/env python

import warnings
from collections import deque, defaultdict
from functools import partial
from pathlib import Path
from typing import Optional

from lhotse import CutSet, load_manifest
from lhotse.utils import Seconds

pzelasko

# from http://stackoverflow.com/questions/569754/how-to-tell-for-which-object-attribute-pickle-fails

"""
Show which fields cannot be pickled
"""

import pickle
def get_pickling_errors(obj,seen=None):
    if seen is None:
        seen = []

pzelasko

#!/usr/bin/env bash

# Common steps
conda create -n k2 python=3.8
conda activate k2 
conda install -c nvidia cudnn=7.6.5 cudatoolkit=10.2
conda install -c pytorch pytorch torchaudio
pip install cmake
mkdir build
pushd build 

pzelasko

import sys
import torch
from torch.utils.data import dataloader
from torch.multiprocessing import reductions
from multiprocessing.reduction import ForkingPickler

default_collate_func = dataloader.default_collate


def default_collate_override(batch):

pzelasko

import numpy as np
import panel as pn
pn.extension()
import holoviews as hv 
hv.notebook_extension("bokeh")
# hv.extension('matplotlib')
from holoviews.streams import Stream, Params
from scipy.io import wavfile
from scipy.signal import spectrogram

pzelasko

sed "/[[:punct:]]*/{ s/[^[:alnum:][:space:]'-]//g}" | tr '[:upper:]' '[:lower:]' | grep -v '[0-9]'

pzelasko

import logging
from sys import stdout

def fancy_logging(level=logging.DEBUG, stream=stdout):
  logging.basicConfig(
    level=level,
    format="[%(asctime)s] %(levelname)s [%(name)s.%(funcName)s:%(lineno)d] %(message)s",
    datefmt="%H:%M:%S",
    stream=stream
  )

pzelasko

#!/bin/bash

host="$1"
source_dir="$2"
target_dir="$3"

ssh "$host" ls "$source_dir" | parallel -j8 -v --sshdelay 0.2 rsync -raz --progress "$host":"$source_dir"/{} "$target_dir"/{}