nikitakrutoy

package main

import (
	"fmt"
	"io/ioutil"
	"log"
	"net/http"
	"time"
)

nikitakrutoy

package main

import (
	"fmt"
	"log"
	"net/http"
)

const TOKEN = "TOKEN"

nikitakrutoy

import graphviz as gv

path = ['Purchase order',
 'Approved',
 'RELEASED (5% delta possible)',
 'Goods Reciept',
 'Invoice Reciept (Logistics invoice) blocked',
 'Invoice Reciept header updated',
 'Invoice Reciept item updated',
 'Invoice Reciept item updated',

nikitakrutoy

                            OLS Regression Results                            
==============================================================================
Dep. Variable:                  price   R-squared:                       0.663
Model:                            OLS   Adj. R-squared:                  0.662
Method:                 Least Squares   F-statistic:                     1307.
Date:                Thu, 24 May 2018   Prob (F-statistic):               0.00
Time:                        01:15:33   Log-Likelihood:                -18811.
No. Observations:                2000   AIC:                         3.763e+04
Df Residuals:                    1997   BIC:                         3.764e+04
Df Model:                           3                                         

nikitakrutoy

OLS Regression Results                            
==============================================================================
Dep. Variable:                  price   R-squared:                       0.668
Model:                            OLS   Adj. R-squared:                  0.667
Method:                 Least Squares   F-statistic:                     2078.
Date:                Thu, 24 May 2018   Prob (F-statistic):               0.00
Time:                        02:44:43   Log-Likelihood:            -2.8798e+06
No. Observations:              319973   AIC:                         5.760e+06
Df Residuals:                  319663   BIC:                         5.764e+06
Df Model:                         309                                         

nikitakrutoy

*Тип продоцва*
* seller_privat -205444.22101121597 
* seller_gewerblich -205673.3093386303

*Тип автомобиля*
vehicleType_cabrio -44260.151805165566
vehicleType_suv -44576.076239830596
vehicleType_coupe -45335.92636625235
vehicleType_bus -45359.941105251564
vehicleType_limousine -45458.52066895459

nikitakrutoy

targets_path = "/Users/nikitakrutoy/playground/ml-practice-homework/hw5/data/svmlin/targets"
objects_path = "/Users/nikitakrutoy/playground/ml-practice-homework/hw5/data/svmlin/objects"

# svmlight data
train_path = "/Users/nikitakrutoy/playground/ml-practice-homework/hw5/data/svmlin/train"

with open(train_path, "r") as f:
    objects_file = open(objects_path, "w")
    targets_file = open(targets_path, "w")
    while True:

nikitakrutoy

using System;
using System.Collections;
using System.Collections.Generic;
using System.IO;

namespace Homework
{
    class Pathway{
        private char _begin;
        private char _end;

nikitakrutoy

reload(hw3code)
print("No cut")
bias, variance = hw3code.compute_bias_variance(DecisionTreeRegressor(), np.sin)
print("Bias %5f Variance %5f" % (bias, variance))
print()

reload(hw3code)
print("Depth 2")
bias, variance = hw3code.compute_bias_variance(DecisionTreeRegressor(max_depth=2), np.sin)
print("Bias %5f Variance %5f" % (bias, variance))

nikitakrutoy

def compute_bias_variance(regressor, dependence_fun, x_generator=np.random.uniform, noise_generator=np.random.uniform,
                          sample_size=300, samples_num=300, objects_num=200, seed=1234):
    """
    После генерации всех необходимых объектов, должна вызываться функция compute_bias_variance_fixed_samples.

    Рекомендации:
    * Создайте вектор объектов для оценивания интеграла по $x$, затем вектор зашумленных правильных ответов.
      Оцените мат. ожидание шума с помощью генерации отдельной шумовой выборки длины objects_num.
    * Проверить правильность реализации можно на примерах, которые разбирались на семинаре и в домашней работе.