AI-2020/raport.md

# Podprojekt Szi

### Opis

Tematem podprojektu jest rozpoznawanie zamówień na podstawie historii zamówień.
Użyłem drzew decyzyjnych.

### Dane

Potrawy, ich nazwa, rodzaj oraz charakterystyka.

    tree_format = ["dish", "served", "price", "origin", "cooked", "ingredients", "name"]

Dane uczące:
    
    dish -  (salad/soup/meal/coffee/tea/non-alcho drink)
    served - (cold/hot/warm)
    origin - (Worldwide/America/Europe/Asia)
    cooked - (baked/boiled/mixed)
    ingridients - (2/4)

Dane testowe jest tworzone losowo w funkcji:  
 
    def client_ordering():
        order = []

        dish = uniq_val_from_data(training_data, 0)
        temperature = uniq_val_from_data(training_data, 1)

        tmpr = random.sample(dish, 1)
        order.append(tmpr[0])

        tmpr = random.sample(temperature, 1)
        order.append(tmpr[0])
        order.append('order')
        return order
  

### Implementacja

#### Drzewo:

Klasy: 
##### Question
    class Queestion:
        def __init__(self, col, value):
            self.col = col      #column
            self.value = value  #value of column
     
        def compare(self, example):
        #compare val in example with val in the question
            
        def __repr__(self):
        #just to print    
        
##### Node
    class Decision_Node():
    #contain the question and child nodes
    def __init__(self, quest, t_branch, f_branch):
        self.quest = quest
        self.t_branch = t_branch
        self.f_branch = f_branch
        
##### Leaf
    class Leaf:
    #contain a number of how many times the label has appeared in dataset
    def __init__(self, rows):
        self.predicts = uniq_count(rows)

 
### Biblioteki

* random
* numpy