Pycaret 3.0 功能抢先体验

点击以下链接可以一键运行例程，无需安装配置环境 ↓ ↓ ↓

Pycaret 3.0 功能抢先体验https://www.heywhale.com/mw/project/62d378a2de3942457a52c905

Pycaret简介

PyCaret 是一个开源、低代码的Python机器学习库，仅用几行代码替换数百行代码，可自动执行机器学习工作流，是一种端到端的机器学习和模型管理工具。

使用手册

官方的使用手册：Welcome to PyCaret - PyCaret Official

支持功能

Pycaret 模块支持的功能有数据处理，模型训练，参数搜索，模型可解释性，模型选择，实验日志查询。

Cheat Sheet

安装Pycaret 3.0 预览版

!pip install llvmlite==0.38.1 -i https://pypi.tuna.tsinghua.edu.cn/simple --ignore-installed
!pip install pycaret==3.0.0rc3 -i https://pypi.tuna.tsinghua.edu.cn/simple #--ignore-installed

查看安装结果

import pycaret
print(pycaret.__version__)

> 3.0.0.rc3

Pycaret 内置数据集

from pycaret.datasets import get_data
all_datasets = get_data('index')
all_datasets

	Dataset	Data Types	Default Task	Target Variable 1	Target Variable 2	# Instances	# Attributes	Missing Values
0	anomaly	Multivariate	Anomaly Detection	None	None	1000	10	N
1	france	Multivariate	Association Rule Mining	InvoiceNo	Description	8557	8	N
2	germany	Multivariate	Association Rule Mining	InvoiceNo	Description	9495	8	N
3	bank	Multivariate	Classification (Binary)	deposit	None	45211	17	N
4	blood	Multivariate	Classification (Binary)	Class	None	748	5	N
5	cancer	Multivariate	Classification (Binary)	Class	None	683	10	N
6	credit	Multivariate	Classification (Binary)	default	None	24000	24	N
7	diabetes	Multivariate	Classification (Binary)	Class variable	None	768	9	N
8	electrical_grid	Multivariate	Classification (Binary)	stabf	None	10000	14	N
9	employee	Multivariate	Classification (Binary)	left	None	14999	10	N
10	heart	Multivariate	Classification (Binary)	DEATH	None	200	16	N
11	heart_disease	Multivariate	Classification (Binary)	Disease	None	270	14	N
12	hepatitis	Multivariate	Classification (Binary)	Class	None	154	32	Y
13	income	Multivariate	Classification (Binary)	income >50K	None	32561	14	Y
14	juice	Multivariate	Classification (Binary)	Purchase	None	1070	15	N
15	nba	Multivariate	Classification (Binary)	TARGET_5Yrs	None	1340	21	N
16	wine	Multivariate	Classification (Binary)	type	None	6498	13	N
17	telescope	Multivariate	Classification (Binary)	Class	None	19020	11	N
18	titanic	Multivariate	Classification (Binary)	Survived	None	891	11	Y
19	us_presidential_election_results	Multivariate	Classification (Binary)	party_winner	None	497	7	N
20	glass	Multivariate	Classification (Multiclass)	Type	None	214	10	N
21	iris	Multivariate	Classification (Multiclass)	species	None	150	5	N
22	poker	Multivariate	Classification (Multiclass)	CLASS	None	100000	11	N
23	questions	Multivariate	Classification (Multiclass)	Next_Question	None	499	4	N
24	satellite	Multivariate	Classification (Multiclass)	Class	None	6435	37	N
25	CTG	Multivariate	Classification (Multiclass)	NSP	None	2129	40	Y
26	asia_gdp	Multivariate	Clustering	None	None	40	11	N
27	elections	Multivariate	Clustering	None	None	3195	54	Y
28	facebook	Multivariate	Clustering	None	None	7050	12	N
29	ipl	Multivariate	Clustering	None	None	153	25	N
30	jewellery	Multivariate	Clustering	None	None	505	4	N
31	mice	Multivariate	Clustering	None	None	1080	82	Y
32	migration	Multivariate	Clustering	None	None	233	12	N
33	perfume	Multivariate	Clustering	None	None	20	29	N
34	pokemon	Multivariate	Clustering	None	None	800	13	Y
35	population	Multivariate	Clustering	None	None	255	56	Y
36	public_health	Multivariate	Clustering	None	None	224	21	N
37	seeds	Multivariate	Clustering	None	None	210	7	N
38	wholesale	Multivariate	Clustering	None	None	440	8	N
39	tweets	Text	NLP	tweet	None	8594	2	N
40	amazon	Text	NLP / Classification	reviewText	None	20000	2	N
41	kiva	Text	NLP / Classification	en	None	6818	7	N
42	spx	Text	NLP / Regression	text	None	874	4	N
43	wikipedia	Text	NLP / Classification	Text	None	500	3	N
44	automobile	Multivariate	Regression	price	None	202	26	Y
45	bike	Multivariate	Regression	cnt	None	17379	15	N
46	boston	Multivariate	Regression	medv	None	506	14	N
47	concrete	Multivariate	Regression	strength	None	1030	9	N
48	diamond	Multivariate	Regression	Price	None	6000	8	N
49	energy	Multivariate	Regression	Heating Load	Cooling Load	768	10	N
50	forest	Multivariate	Regression	area	None	517	13	N
51	gold	Multivariate	Regression	Gold_T+22	None	2558	121	N
52	house	Multivariate	Regression	SalePrice	None	1461	81	Y
53	insurance	Multivariate	Regression	charges	None	1338	7	N
54	parkinsons	Multivariate	Regression	PPE	None	5875	22	N
55	traffic	Multivariate	Regression	traffic_volume	None	48204	8	N

>

	Dataset	Data Types	Default Task	Target Variable 1	Target Variable 2	# Instances	# Attributes	Missing Values
0	anomaly	Multivariate	Anomaly Detection	None	None	1000	10	N
1	france	Multivariate	Association Rule Mining	InvoiceNo	Description	8557	8	N
2	germany	Multivariate	Association Rule Mining	InvoiceNo	Description	9495	8	N
3	bank	Multivariate	Classification (Binary)	deposit	None	45211	17	N
4	blood	Multivariate	Classification (Binary)	Class	None	748	5	N
5	cancer	Multivariate	Classification (Binary)	Class	None	683	10	N
6	credit	Multivariate	Classification (Binary)	default	None	24000	24	N
7	diabetes	Multivariate	Classification (Binary)	Class variable	None	768	9	N
8	electrical_grid	Multivariate	Classification (Binary)	stabf	None	10000	14	N
9	employee	Multivariate	Classification (Binary)	left	None	14999	10	N
10	heart	Multivariate	Classification (Binary)	DEATH	None	200	16	N
11	heart_disease	Multivariate	Classification (Binary)	Disease	None	270	14	N
12	hepatitis	Multivariate	Classification (Binary)	Class	None	154	32	Y
13	income	Multivariate	Classification (Binary)	income >50K	None	32561	14	Y
14	juice	Multivariate	Classification (Binary)	Purchase	None	1070	15	N
15	nba	Multivariate	Classification (Binary)	TARGET_5Yrs	None	1340	21	N
16	wine	Multivariate	Classification (Binary)	type	None	6498	13	N
17	telescope	Multivariate	Classification (Binary)	Class	None	19020	11	N
18	titanic	Multivariate	Classification (Binary)	Survived	None	891	11	Y
19	us_presidential_election_results	Multivariate	Classification (Binary)	party_winner	None	497	7	N
20	glass	Multivariate	Classification (Multiclass)	Type	None	214	10	N
21	iris	Multivariate	Classification (Multiclass)	species	None	150	5	N
22	poker	Multivariate	Classification (Multiclass)	CLASS	None	100000	11	N
23	questions	Multivariate	Classification (Multiclass)	Next_Question	None	499	4	N
24	satellite	Multivariate	Classification (Multiclass)	Class	None	6435	37	N
25	CTG	Multivariate	Classification (Multiclass)	NSP	None	2129	40	Y
26	asia_gdp	Multivariate	Clustering	None	None	40	11	N
27	elections	Multivariate	Clustering	None	None	3195	54	Y
28	facebook	Multivariate	Clustering	None	None	7050	12	N
29	ipl	Multivariate	Clustering	None	None	153	25	N
30	jewellery	Multivariate	Clustering	None	None	505	4	N
31	mice	Multivariate	Clustering	None	None	1080	82	Y
32	migration	Multivariate	Clustering	None	None	233	12	N
33	perfume	Multivariate	Clustering	None	None	20	29	N
34	pokemon	Multivariate	Clustering	None	None	800	13	Y
35	population	Multivariate	Clustering	None	None	255	56	Y
36	public_health	Multivariate	Clustering	None	None	224	21	N
37	seeds	Multivariate	Clustering	None	None	210	7	N
38	wholesale	Multivariate	Clustering	None	None	440	8	N
39	tweets	Text	NLP	tweet	None	8594	2	N
40	amazon	Text	NLP / Classification	reviewText	None	20000	2	N
41	kiva	Text	NLP / Classification	en	None	6818	7	N
42	spx	Text	NLP / Regression	text	None	874	4	N
43	wikipedia	Text	NLP / Classification	Text	None	500	3	N
44	automobile	Multivariate	Regression	price	None	202	26	Y
45	bike	Multivariate	Regression	cnt	None	17379	15	N
46	boston	Multivariate	Regression	medv	None	506	14	N
47	concrete	Multivariate	Regression	strength	None	1030	9	N
48	diamond	Multivariate	Regression	Price	None	6000	8	N
49	energy	Multivariate	Regression	Heating Load	Cooling Load	768	10	N
50	forest	Multivariate	Regression	area	None	517	13	N
51	gold	Multivariate	Regression	Gold_T+22	None	2558	121	N
52	house	Multivariate	Regression	SalePrice	None	1461	81	Y
53	insurance	Multivariate	Regression	charges	None	1338	7	N
54	parkinsons	Multivariate	Regression	PPE	None	5875	22	N
55	traffic	Multivariate	Regression	traffic_volume	None	48204	8	N

Pycaret 时间序列预测

导入模块

# 导入Pycaret内置数据
from pycaret.datasets import get_data
# 导入Pycaret时间序列预测模型（3.x版本新增）
from pycaret.time_series import *

import matplotlib.pyplot as plt

import warnings
warnings.filterwarnings('ignore')

读取数据

# 使用内置数据airline
data = get_data('airline')
data

> 

Period
1949-01    112.0
1949-02    118.0
1949-03    132.0
1949-04    129.0
1949-05    121.0
Freq: M, Name: Number of airline passengers, dtype: float64

>

Period
1949-01    112.0
1949-02    118.0
1949-03    132.0
1949-04    129.0
1949-05    121.0
           ...  
1960-08    606.0
1960-09    508.0
1960-10    461.0
1960-11    390.0
1960-12    432.0
Freq: M, Name: Number of airline passengers, Length: 144, dtype: float64

数据探索分析

# 时间序列绘制
data.plot()

>

# 统计测试
check_stats()

>

	Test	Test Name	Data	Property	Setting	Value
0	Summary	Statistics	Transformed	Length		144.0
1	Summary	Statistics	Transformed	# Missing Values		0.0
2	Summary	Statistics	Transformed	Mean		280.298611
3	Summary	Statistics	Transformed	Median		265.5
4	Summary	Statistics	Transformed	Standard Deviation		119.966317
5	Summary	Statistics	Transformed	Variance		14391.917201
6	Summary	Statistics	Transformed	Kurtosis		-0.364942
7	Summary	Statistics	Transformed	Skewness		0.58316
8	Summary	Statistics	Transformed	# Distinct Values		118.0
9	White Noise	Ljung-Box	Transformed	Test Statictic	{'alpha': 0.05, 'K': 24}	1606.083817
10	White Noise	Ljung-Box	Transformed	Test Statictic	{'alpha': 0.05, 'K': 48}	1933.155822
11	White Noise	Ljung-Box	Transformed	p-value	{'alpha': 0.05, 'K': 24}	0.0
12	White Noise	Ljung-Box	Transformed	p-value	{'alpha': 0.05, 'K': 48}	0.0
13	White Noise	Ljung-Box	Transformed	White Noise	{'alpha': 0.05, 'K': 24}	False
14	White Noise	Ljung-Box	Transformed	White Noise	{'alpha': 0.05, 'K': 48}	False
15	Stationarity	ADF	Transformed	Stationarity	{'alpha': 0.05}	False
16	Stationarity	ADF	Transformed	p-value	{'alpha': 0.05}	0.99188
17	Stationarity	ADF	Transformed	Test Statistic	{'alpha': 0.05}	0.815369
18	Stationarity	ADF	Transformed	Critical Value 1%	{'alpha': 0.05}	-3.481682
19	Stationarity	ADF	Transformed	Critical Value 5%	{'alpha': 0.05}	-2.884042
20	Stationarity	ADF	Transformed	Critical Value 10%	{'alpha': 0.05}	-2.57877
21	Stationarity	KPSS	Transformed	Trend Stationarity	{'alpha': 0.05}	True
22	Stationarity	KPSS	Transformed	p-value	{'alpha': 0.05}	0.1
23	Stationarity	KPSS	Transformed	Test Statistic	{'alpha': 0.05}	0.09615
24	Stationarity	KPSS	Transformed	Critical Value 10%	{'alpha': 0.05}	0.119
25	Stationarity	KPSS	Transformed	Critical Value 5%	{'alpha': 0.05}	0.146
26	Stationarity	KPSS	Transformed	Critical Value 2.5%	{'alpha': 0.05}	0.176
27	Stationarity	KPSS	Transformed	Critical Value 1%	{'alpha': 0.05}	0.216
28	Normality	Shapiro	Transformed	Normality	{'alpha': 0.05}	False
29	Normality	Shapiro	Transformed	p-value	{'alpha': 0.05}	0.000068

寻找最佳模型

# 初始化
s = setup(data, fh = 12, session_id = 123) 
# 模型比较
best = compare_models()

> 

	Description	Value
0	session_id	123
1	Target	Number of airline passengers
2	Approach	Univariate
3	Exogenous Variables	Not Present
4	Original data shape	(144, 1)
5	Transformed data shape	(144, 1)
6	Transformed train set shape	(132, 1)
7	Transformed test set shape	(12, 1)
8	Rows with missing values	0.0%
9	Fold Generator	ExpandingWindowSplitter
10	Fold Number	3
11	Enforce Prediction Interval	False
12	Seasonal Period(s) Tested	12
13	Seasonality Present	True
14	Seasonalities Detected	[12]
15	Primary Seasonality	12
16	Target Strictly Positive	True
17	Target White Noise	No
18	Recommended d	1
19	Recommended Seasonal D	1
20	Preprocess	False
21	CPU Jobs	-1
22	Use GPU	False
23	Log Experiment	False
24	Experiment Name	ts-default-name
25	USI	98d2

	Model	MASE	RMSSE	MAE	RMSE	MAPE	SMAPE	R2	TT (Sec)
exp_smooth	Exponential Smoothing	0.5716	0.5997	16.7767	19.7954	0.0422	0.0427	0.8954	0.0400
ets	ETS	0.5931	0.6212	17.4172	20.5108	0.0440	0.0445	0.8882	0.0767
et_cds_dt	Extra Trees w/ Cond. Deseasonalize & Detrending	0.6602	0.7288	19.4653	24.1050	0.0484	0.0484	0.8459	0.1167
huber_cds_dt	Huber w/ Cond. Deseasonalize & Detrending	0.6813	0.7866	20.0334	25.9670	0.0491	0.0499	0.8113	0.0267
arima	ARIMA	0.6830	0.6735	20.0069	22.2199	0.0501	0.0507	0.8677	0.0900
lr_cds_dt	Linear w/ Cond. Deseasonalize & Detrending	0.7004	0.7702	20.6084	25.4401	0.0509	0.0514	0.8215	0.0267
ridge_cds_dt	Ridge w/ Cond. Deseasonalize & Detrending	0.7004	0.7703	20.6086	25.4405	0.0509	0.0514	0.8215	0.0233
lar_cds_dt	Least Angular Regressor w/ Cond. Deseasonalize & Detrending	0.7004	0.7702	20.6084	25.4401	0.0509	0.0514	0.8215	0.0233
en_cds_dt	Elastic Net w/ Cond. Deseasonalize & Detrending	0.7029	0.7732	20.6816	25.5362	0.0511	0.0516	0.8201	0.0267
lasso_cds_dt	Lasso w/ Cond. Deseasonalize & Detrending	0.7048	0.7751	20.7373	25.6005	0.0512	0.0517	0.8193	0.0200
catboost_cds_dt	CatBoost Regressor w/ Cond. Deseasonalize & Detrending	0.7106	0.8146	20.9112	26.8907	0.0505	0.0509	0.8085	0.9433
br_cds_dt	Bayesian Ridge w/ Cond. Deseasonalize & Detrending	0.7112	0.7837	20.9213	25.8795	0.0515	0.0521	0.8144	0.0233
knn_cds_dt	K Neighbors w/ Cond. Deseasonalize & Detrending	0.7162	0.8157	21.1613	26.9700	0.0521	0.0529	0.7811	0.0300
auto_arima	Auto ARIMA	0.7181	0.7114	21.0297	23.4661	0.0525	0.0531	0.8509	1.6967
gbr_cds_dt	Gradient Boosting w/ Cond. Deseasonalize & Detrending	0.7938	0.9310	23.3723	30.7344	0.0569	0.0576	0.7417	0.0367
xgboost_cds_dt	Extreme Gradient Boosting w/ Cond. Deseasonalize & Detrending	0.8155	0.9591	24.0738	31.6950	0.0582	0.0592	0.7118	152.0800
lightgbm_cds_dt	Light Gradient Boosting w/ Cond. Deseasonalize & Detrending	0.8156	0.9117	24.0002	30.0956	0.0575	0.0587	0.7561	86.6467
rf_cds_dt	Random Forest w/ Cond. Deseasonalize & Detrending	0.8327	0.9465	24.5290	31.2635	0.0600	0.0606	0.7360	0.1400
ada_cds_dt	AdaBoost w/ Cond. Deseasonalize & Detrending	0.8825	1.0292	25.9471	33.9304	0.0619	0.0637	0.6725	0.0500
llar_cds_dt	Lasso Least Angular Regressor w/ Cond. Deseasonalize & Detrending	0.9670	1.1915	28.4499	39.3303	0.0665	0.0693	0.5738	0.0233
theta	Theta Forecaster	0.9729	1.0306	28.3192	33.8639	0.0670	0.0700	0.6710	0.0167
omp_cds_dt	Orthogonal Matching Pursuit w/ Cond. Deseasonalize & Detrending	1.0090	1.2370	29.6294	40.8121	0.0685	0.0718	0.5462	0.0200
dt_cds_dt	Decision Tree w/ Cond. Deseasonalize & Detrending	1.0429	1.2226	30.4800	40.1912	0.0726	0.0753	0.5362	0.0333
snaive	Seasonal Naive Forecaster	1.1479	1.0945	33.3611	35.9139	0.0832	0.0879	0.6072	0.0133
par_cds_dt	Passive Aggressive w/ Cond. Deseasonalize & Detrending	1.2472	1.3081	36.7727	43.3215	0.0935	0.0961	0.4968	0.0233
polytrend	Polynomial Trend Forecaster	1.6523	1.9202	48.6301	63.4299	0.1170	0.1216	-0.0784	0.0100
croston	Croston	1.9311	2.3517	56.6180	77.5856	0.1295	0.1439	-0.6281	0.0100
naive	Naive Forecaster	2.3599	2.7612	69.0278	91.0322	0.1569	0.1792	-1.2216	0.6467
grand_means	Grand Means Forecaster	5.5306	5.2596	162.4117	173.6492	0.4000	0.5075	-7.0462	0.5100

Processing:   0%|          | 0/125 [00:00<?, ?it/s]

# 查看最优模型
best

>

ExponentialSmoothing

ExponentialSmoothing(seasonal='mul', sp=12, trend='add')

exp_smooth = create_model('exp_smooth')
print(exp_smooth)

>

	cutoff	MASE	RMSSE	MAE	RMSE	MAPE	SMAPE	R2
0	1956-12	0.4985	0.5735	14.5584	18.7730	0.0366	0.0376	0.8853
1	1957-12	0.5088	0.5368	15.5548	18.2243	0.0420	0.0411	0.9130
2	1958-12	0.7075	0.6888	20.2167	22.3888	0.0479	0.0494	0.8879
Mean	NaT	0.5716	0.5997	16.7767	19.7954	0.0422	0.0427	0.8954
SD	NaT	0.0962	0.0647	2.4663	1.8475	0.0046	0.0049	0.0125

Processing:   0%|          | 0/4 [00:00<?, ?it/s]

ExponentialSmoothing(seasonal='mul', sp=12, trend='add')

# 模型预测评估
pred = predict_model(exp_smooth)

>

	Model	MASE	RMSSE	MAE	RMSE	MAPE	SMAPE	R2
0	Exponential Smoothing	0.3383	0.4576	10.3023	15.8096	0.0221	0.0216	0.9549