sneakers-the-rat

require(stats)
require(ggplot2)

# Download your last.fm history as a .csv from here: 
# https://benjaminbenben.com/lastfm-to-csv/

# load 'em
lfmdat <- read.csv('~/lastfm.csv', header = FALSE)
names(lfmdat) <- c("artist", "album", "song", "time")

sneakers-the-rat

## in model.load_model:
# add import
from .utils.io_utils import ask_to_proceed_with_overwrite, update_config

# At 225 add a version check and call an updating function
model_config = json.loads(model_config.decode('utf-8'))

k_version = float(f.attrs.get('keras_version')[0])
if k_version < 2:
    # Have to rename params to load old models

sneakers-the-rat

import os
import pandas as pd
import numpy as np
from itertools import cycle
from tqdm import tqdm
from multiprocessing import Pool
import time

def concat_sport(sport, i):
    sport_files = [s for s in files if s.startswith(sport)]

sneakers-the-rat

# this should let ya surf the line.
# take edge coordinates (1 px thick) and reorder 
# so first-last goes from one end of the line to the other

import numpy as np
from skimage import morphology
from scipy.spatial import distance
from collections import deque as dq

def order_edges(edges):

sneakers-the-rat

# a more or less line-for-line transcription of the source available here:
# https://docs.google.com/open?id=0B10RxHxW3I92dG9SU0pNMV84alk
# which is described in this paper:
# http://ieeexplore.ieee.org/document/6166585/
# D. K. Prasad, C. Quek, M. K. H. Leung and S. Y. Cho, "A parameter independent line fitting method," The First Asian Conference on Pattern Recognition, Beijing, 2011, pp. 441-445.
# doi: 10.1109/ACPR.2011.6166585
#
#
# Operation is pretty simple, just pass an **ordered** Nx2 array of x/y coordinates, get line segments.
# A convenience ordering function is provided free of charge ;)

sneakers-the-rat

library(ggplot2)

jobs <- data.frame(start=c(2000, 2006, 2008, 2009, 2014),
                   end  =c(2005, 2009, 2010, 2015, 2015),
                   idx = rev(seq(5)),
                   name =as.ordered(c("job a", 'job b', 'job c', 'job d', 'job e')))

g.job <- ggplot(jobs)+
  geom_rect(aes(xmin=start, xmax=end, ymin=idx, ymax=idx+1, fill=name))+
  scale_fill_grey(start=0, end=0.8)+

sneakers-the-rat

# renamed files to 'vid.avi' and 'points.csv'
# deleted first row of points.csv

# call this and use -p 'y'/'n', -r 'y'/'n' to control playing and rendering the video
# or do ./elliot_pupil.py --help

# you could comment out the skvideo and tqdm imports and objects if ya don't want to install/render.
import cv2
import pandas as pd
import numpy as np

sneakers-the-rat

<scheme name="Peacocks In Space Contrast (rainglow)" version="142" parent_scheme="Darcula">
  <option name="FONT_SCALE" value="1.0" />
  <metaInfo>
    <property name="created">2019-01-03T19:28:35</property>
    <property name="ide">PyCharmCore</property>
    <property name="ideVersion">2018.2.0.0</property>
    <property name="modified">2019-01-03T19:28:43</property>
    <property name="originalScheme">_@user_Peacocks In Space Contrast (rainglow)</property>
  </metaInfo>
  <option name="LINE_SPACING" value="1.4" />

sneakers-the-rat

library(rvest)
library(dplyr)
library(stringr)

m200 <- read_html("http://www.alltime-athletics.com/m_100ok.htm")

m2_table <- m200 %>%
  html_nodes("body > center:nth-child(8) > pre")

# make list of rows, split by newline character

sneakers-the-rat

library(psych)
library(ggplot2)
library(plyr)
library(lme4)
library(lmerTest)

n_sims <- 1000

sims <- expand.grid(
  i_sim     =  seq(n_sims),