用jupyter notebook寫的第一個機器學習

模型：線性函式
演算法：梯度下降

import torch

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

data = pd.read_csv('dataset/Income1.csv') # 匯入資料

data

	Unnamed: 0	Education	Income
0	1	10.000000	26.658839
1	2	10.401338	27.306435
2	3	10.842809	22.132410
3	4	11.244147	21.169841
4	5	11.645485	15.192634
5	6	12.086957	26.398951
6	7	12.488294	17.435307
7	8	12.889632	25.507885
8	9	13.290970	36.884595
9	10	13.732441	39.666109
10	11	14.133779	34.396281
11	12	14.535117	41.497994
12	13	14.976589	44.981575
13	14	15.377926	47.039595
14	15	15.779264	48.252578
15	16	16.220736	57.034251
16	17	16.622074	51.490919
17	18	17.023411	61.336621
18	19	17.464883	57.581988
19	20	17.866221	68.553714
20	21	18.267559	64.310925
21	22	18.709030	68.959009
22	23	19.110368	74.614639
23	24	19.511706	71.867195
24	25	19.913043	76.098135
25	26	20.354515	75.775218
26	27	20.755853	72.486055
27	28	21.157191	77.355021
28	29	21.598662	72.118790
29	30	22.000000	80.260571

data.info()  # 檢視資料資訊

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 30 entries, 0 to 29
Data columns (total 3 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   Unnamed: 0  30 non-null     int64  
 1   Education   30 non-null     float64
 2   Income      30 non-null     float64
dtypes: float64(2), int64(1)
memory usage: 848.0 bytes

畫散點圖

plt.scatter(data.Education, data.Income)
plt.xlabel('education')
plt.ylabel('income')
plt.show()

from torch import nn

X = torch.from_numpy(data.Education.values.reshape(-1, 1).astype(np.float32))

從data.Education中提取value，並重新把它變成只有1列的矩陣，同時把資料型別轉化為np.float32，然後把這個numpy型別轉換成torch型別

X

tensor([[10.0000],
        [10.4013],
        [10.8428],
        [11.2441],
        [11.6455],
        [12.0870],
        [12.4883],
        [12.8896],
        [13.2910],
        [13.7324],
        [14.1338],
        [14.5351],
        [14.9766],
        [15.3779],
        [15.7793],
        [16.2207],
        [16.6221],
        [17.0234],
        [17.4649],
        [17.8662],
        [18.2676],
        [18.7090],
        [19.1104],
        [19.5117],
        [19.9130],
        [20.3545],
        [20.7559],
        [21.1572],
        [21.5987],
        [22.0000]])

X.shape

torch.Size([30, 1])

Y = torch.from_numpy(data.Income.values.reshape(-1, 1).astype(np.float32))

Y

tensor([[26.6588],
        [27.3064],
        [22.1324],
        [21.1698],
        [15.1926],
        [26.3990],
        [17.4353],
        [25.5079],
        [36.8846],
        [39.6661],
        [34.3963],
        [41.4980],
        [44.9816],
        [47.0396],
        [48.2526],
        [57.0343],
        [51.4909],
        [61.3366],
        [57.5820],
        [68.5537],
        [64.3109],
        [68.9590],
        [74.6146],
        [71.8672],
        [76.0981],
        [75.7752],
        [72.4861],
        [77.3550],
        [72.1188],
        [80.2606]])

Y.shape

torch.Size([30, 1])

建立模型

model = nn.Linear(1, 1)  # output = w @ input + b , output和input長度為1 @代表矩陣乘法

loss_fn = nn.MSELoss()   # 建立損失函式

opt = torch.optim.SGD(model.parameters(), lr=0.0001) # 優化的設定

for epoch in range(5000):       # 訓練全部資料
    for x,y in zip(X,Y):
        y_pred = model(x)       # 使用模型預測
        loss = loss_fn(y, y_pred) # 計算損失
        opt.zero_grad()            # 梯度清零
        loss.backward()            # 計算梯度
        opt.step()                 # 優化模型引數w和b

model.weight  # 檢視訓練完成的w

Parameter containing:
tensor([[3.4113]], requires_grad=True)

model.bias    # 檢視訓練完成的b

Parameter containing:
tensor([0.4944], requires_grad=True)

結果視覺化

plt.scatter(data.Education, data.Income)
plt.plot(X.numpy(), model(X).data.numpy(), c='r')  # .numpy()是把tensor轉化成numpy中的array型別
plt.show()