Dmitry Petukhov codez0mb1e

Directories:

.                                         <- project root
|-- src/                                  <- source codes
|   |-- middleware/                       <- API middleware
|       |-- web3.py
|       |-- auth.py
|       |-- errors.py
|       |-- ...

Template for Data Science Project

Main Principles

Reproducibility:
- code files: under version control, code review
- data: data pipeline or snapshots
- environment: venv/conda/docker
- models: training pipeline or pickled models, saved hyper-parameters and metrics
- experiment: tracking, report

	# %% errors.py ----
	from abc import ABC
	from dataclasses import dataclass
	from enum import Enum, unique


	class BaseWalletError(Exception, ABC):
	"""Base exception for wallet-level errors"""
	pass

	#!/usr/bin/python3

	""" [Description]
	"""

	# core
	import os
	import gc

	# data science

	USE <db_name>
	GO

	CREATE TABLE <schema_name>.<table_name>(
	id int,
	name varchar(255),
	...
	) ON [PRIMARY]
	GO


	#'
	#' ML Experiment functions
	#'



	#' Get experiment root directory
	#'
	#' @param .config

	default:
	# Common ----
	experiment_dir: '.' # relative from project directory as root
	data_dir: '../data' # relative from @experiment_dir as root
	cache_dir: 'cache' # relative from @experiment_dir as root
	output_dir: 'output' # relative from @experiment_dir as root
	experiment_version: '2020-01-01'

	stages:
	# Downloading data stage config ----

	#'
	#' Sanitize character columns in dataframe
	#'
	#' @param dt
	#' @param .extra_columns
	#' @param .algo
	#' @param .verbose
	#'
	sanitize_dataframe <- function(dt,
	.extra_columns = NULL,

	#'
	#' IO functions wrapper
	#'


	#' Write dataset to CSV
	#'
	#' @param dt
	#' @param .file_name
	#' @param .col_names

	WITH duplicates AS (
	SELECT
	*,
	ROW_NUMBER() OVER (
	PARTITION BY Id
	ORDER BY [Timestamp]
	) row_n
	FROM
	<table_name> -- NOTE: insert name of target table here
	WHERE