In JohnGavin/ccxt: ccxt Analysis

https://www.r-bloggers.com/2022/02/r-python-rosetta-stone-reading-files-and-column-transformations

libs <- c('dplyr', 'tidyr', 'stringr',  # wrangling
          'knitr','kableExtra',         # table styling
          'ggplot2','gridExtra',        # plots
          'viridis',                    # visuals styling
          'reticulate')                 # e pluribus unum
# devtools::install_cran(c('pacman', libs))
# pacman::p_load(libs)
invisible(lapply(libs, library, character.only = TRUE))

install.packages(c("gt", "palmerpenguins", 'vroom',
  'readr'))

libs <- c('dplyr', 'stringr',     # wrangling
          'palmerpenguins', 'gt', # data, table styling
          'vroom', 'readr',       # read & write data
          'tidyr', 'purrr',       # wrangle & iterate
          'fs',                   # file system
          'reticulate')           # python support 
invisible(lapply(libs, library, character.only = TRUE))

# use_python("/usr/bin/python")
use_python("/usr/bin/python3")
df_orig <- penguins %>% 
  mutate_if(is.integer, as.double) %>% 
  select(-contains("length"), -contains("depth"))

import pandas as pd
import glob
pd.set_option('display.max_columns', None)

df_orig %>% 
  count(species) %>% 
  gt()

fp <- "data/penguins/"
df_orig %>%
  group_by(species) %>% 
  # slicing off the first n entries (here by group
  slice_head(n = 1) %>% 
  ungroup() %>% 
  nest(data = c(-species)) %>% 
  pwalk(function(species, data) 
    write_csv(data, 
      file.path(fp, str_c(species, "_penguins.csv"))))

single_file <- vroom(
  paste0(fp, "Adelie_penguins.csv", col_types = cols()))
single_file %>% 
  gt()

single_file = pd.read_csv(r.fp + "Adelie_penguins.csv")
single_file

(files <- fs::dir_ls(fp, glob = "*penguins.csv"))
## Adelie_penguins.csv    Chinstrap_penguins.csv Gentoo_penguins.csv
df <- vroom(files, col_types = cols(), id = "name")
df %>% 
  gt()

In Python, we use glob to grab the file names:

files_py = glob.glob(r.fp + "*penguins.csv")
files_py
# df_py = pd.concat((pd.read_csv(f) for f in files_py))
# file name. Luckily, we can get that information through the nifty assign function:
df_py = pd.concat((pd.read_csv(f).assign(name = f) for f in files_py))
df_py

df_py

df %>% 
  separate(name, into = c("name", "filetype"), sep = "\\.") %>% 
  gt()

df %>% 
  separate(name, into = c("name", "filetype"), sep = "\\.") %>% 
  separate(name, into = c("species", "animal"), sep = "_") %>% 
  gt()

df_py = pd.concat((pd.read_csv(f).assign(name = f) for f in files_py))
df_py[['name', 'filetype']] = 
  df_py['name'].str.split('\\.', expand=True)
df_py[['species', 'animal']] = df_py['name'].str.split('_', expand=True)
df_py = df_py.drop('name', axis = 'columns')
df_py

df %>% 
  separate(name, into = c("name", "filetype"), sep = "\\.") %>% 
  separate(name, into = c("species", "animal"), sep = "_") %>% 
  gt()

df %>% 
  separate(name, into = c("name", "filetype"), sep = "\\.") %>% 
  separate(name, into = c("species", "animal"), sep = "_") %>% 
  unite(species, animal, col = "name", sep = "_") %>% 
  unite(name, filetype, col = "name", sep = ".") %>% 
  gt()

df_py['name'] = df_py['species'] + "_" + df_py['animal'] + "." + df_py['filetype']
df_py = df_py.drop(['species', 'animal', 'filetype'], axis = 'columns')
df_py

foo = [1, 2, 3]
print(foo[0])
print(r.iris.loc[:5, ["Sepal.Length", "Species"]])

import pandas
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier

train, test = train_test_split(r.iris,
                test_size = 0.4, random_state = 4321)

X = train.drop('Species', axis = 1)
y = train.loc[:, 'Species'].values
X_test = test.drop('Species', axis = 1)
y_test = test.loc[:, 'Species'].values

tree = DecisionTreeClassifier(random_state=4321)
clf = tree.fit(X, y)
pred = clf.predict(X_test)

JohnGavin/ccxt documentation built on Sept. 3, 2022, 5:53 a.m.