harshvardhaniimi

Calibre App Enhancements

• Calibre Plugins: Enhance your Calibre app with a variety of plugins available at GitHub - Calibre Plugins by a great developer, or explore the complete directory.
  • K2pdfopt: Converts PDF to a format readable on Kindle. More information can be found on MobileRead Forums.
  • Page Count and Readability Indexes: Adds a page count column and readability indexes to the Calibre app. Details are available on GitHub - Count Pages Wiki.

Kindle Highlights Management

• Emdash: An AI-powered wisdom indexer that organizes text snippets for better recall and learning, functioning completely offline. Open-source and free. GitHub - Emdash.
• Remind Kindle: Sends reminder emails from your Kindle highlights using Clippings.t

harshvardhaniimi

# pip install lightgbm optuna

# Also read:
# 1. https://optuna.org/
# 2. https://archive.is/sYEBT
# 3. https://archive.is/lrcTV

import optuna
import lightgbm as lgb
from sklearn.metrics import mean_squared_error

harshvardhaniimi

import pandas as pd
import re

def clean_names(df):
    """
    This function takes in a pandas dataframe, and renames its columns by removing special characters 
    and spaces, replacing them with underscores. It also converts all letters to lowercase.
    """
    df.columns = df.columns.str.lower()  # Convert to lowercase
    df.columns = df.columns.str.replace(' ', '_', regex=False)  # Replace spaces with underscores

harshvardhaniimi

import pandas as pd

def get_categorical_uniques(df, cols=None):
    """
    This function identifies categorical columns as those with the 'object' or 'category' data type. 
    If a list of columns is provided, it will only consider those columns. The output DataFrame 
    has three columns: 'Column Name' for the name of the input column, 'Number of Unique Values' 
    for the count of unique values in the input column, and 'Unique Values' for a list of the 
    unique values.

harshvardhaniimi

import pandas as pd
import numpy as np

def describe_dataframe(df):
    """
    Describes a dataframe in terms of unique values of categorical variables,
    range of continuous variables and proportion of missing values.
    """
    results = []
    for column in df.columns:

harshvardhaniimi

import pandas as pd
import hashlib

def hash_dataframe(df):
    """
    Generate a hash for a DataFrame using the SHA-256 algorithm.
    
    This function creates a hash for each row of the DataFrame using pandas' `hash_pandas_object` 
    and then hashes the resulting array of row hashes using `hashlib.sha256`.
    

harshvardhaniimi

# Citation: https://imgflip.com/i/7im94j
import pandas as pd
import numpy as np

def downcast_dataframe(df):
    """
    Downcasts a pandas DataFrame's columns to the minimum resolution required, modifying it in place.
    It retains categorical columns and prints the changes made along with the reduction in memory consumption.

    Parameters

harshvardhaniimi

library(tidyverse)
library(readxl)
library(MetBrewer)
theme_set(ggthemes::theme_clean())

setwd("/Users/harshvardhan/Desktop/Dump/diapers/")

# its funny how something so simple requires such ninja skills
# 1. Data from Statista is sometimes in millions, sometimes in billions
# 2. You cannot download more than three countries of data at a time

harshvardhaniimi

harshvardhaniimi

# Install a conda package in the current Jupyter kernel
import sys
!conda install --yes --prefix {sys.prefix} numpy


# Install a pip package in the current Jupyter kernel
import sys
!{sys.executable} -m pip install numpy

# Learn more: https://jakevdp.github.io/blog/2017/12/05/installing-python-packages-from-jupyter/