AVOCADO 가격 예측 (Facebook Prophet )¶

STEP #0: 데이터셋¶

데이터는 미국의 아보카도 리테일 데이터 입니다. (2018년도 weekly 데이터)
아보카도 거래량과 가격이 나와 있습니다.

컬럼 설명 :

Date - The date of the observation
AveragePrice - the average price of a single avocado
type - conventional or organic
year - the year
Region - the city or region of the observation
Total Volume - Total number of avocados sold
4046 - Total number of avocados with PLU 4046 sold - PLU는 농산물 코드입니다
4225 - Total number of avocados with PLU 4225 sold
4770 - Total number of avocados with PLU 4770 sold

STEP #1: 데이터 준비¶

Prophet 라이브러리¶

install : pip install fbprophet
위 에러 발생시 : conda install -c conda-forge fbprophet
레퍼런스 : https://research.fb.com/prophet-forecasting-at-scale/

https://facebook.github.io/prophet/docs/quick_start.html#python-api

In [ ]:

# 프로펫 라이브러리가 fbprophet 에서 그냥 prophet 으로 변경되었음.

In [4]:

# import libraries 
import pandas as pd  
import numpy as np 
import matplotlib.pyplot as plt 
import random
import seaborn as sns
from prophet import Prophet

In [5]:

from google.colab import drive
drive.mount('/content/drive')

Mounted at /content/drive

In [6]:

# 워킹디렉토리 설정
import os 
os.chdir('/content/drive/MyDrive/Colab Notebooks/ml_plus/data')

In [9]:

# avocado.csv 데이터 읽기
df =  pd.read_csv('avocado.csv', index_col = 0) # index_col = 0 (첫번째 컬럼을 인덱스로 사용하라.)

In [ ]:

# 날짜별로 데이터가 있는 형식 => Time Series Data

In [10]:

df

Out[10]:

	Date	AveragePrice	Total Volume	4046	4225	4770	Total Bags	Small Bags	Large Bags	XLarge Bags	type	year	region
0	2015-12-27	1.33	64236.62	1036.74	54454.85	48.16	8696.87	8603.62	93.25	0.0	conventional	2015	Albany
1	2015-12-20	1.35	54876.98	674.28	44638.81	58.33	9505.56	9408.07	97.49	0.0	conventional	2015	Albany
2	2015-12-13	0.93	118220.22	794.70	109149.67	130.50	8145.35	8042.21	103.14	0.0	conventional	2015	Albany
3	2015-12-06	1.08	78992.15	1132.00	71976.41	72.58	5811.16	5677.40	133.76	0.0	conventional	2015	Albany
4	2015-11-29	1.28	51039.60	941.48	43838.39	75.78	6183.95	5986.26	197.69	0.0	conventional	2015	Albany
...	...	...	...	...	...	...	...	...	...	...	...	...	...
7	2018-02-04	1.63	17074.83	2046.96	1529.20	0.00	13498.67	13066.82	431.85	0.0	organic	2018	WestTexNewMexico
8	2018-01-28	1.71	13888.04	1191.70	3431.50	0.00	9264.84	8940.04	324.80	0.0	organic	2018	WestTexNewMexico
9	2018-01-21	1.87	13766.76	1191.92	2452.79	727.94	9394.11	9351.80	42.31	0.0	organic	2018	WestTexNewMexico
10	2018-01-14	1.93	16205.22	1527.63	2981.04	727.01	10969.54	10919.54	50.00	0.0	organic	2018	WestTexNewMexico
11	2018-01-07	1.62	17489.58	2894.77	2356.13	224.53	12014.15	11988.14	26.01	0.0	organic	2018	WestTexNewMexico

18249 rows × 13 columns

In [ ]:

STEP #2: EDA(Exploratory Data Analysis) : 탐색적 데이터 분석¶

In [11]:

df.describe()

Out[11]:

	AveragePrice	Total Volume	4046	4225	4770	Total Bags	Small Bags	Large Bags	XLarge Bags	year
count	18249.000000	1.824900e+04	1.824900e+04	1.824900e+04	1.824900e+04	1.824900e+04	1.824900e+04	1.824900e+04	18249.000000	18249.000000
mean	1.405978	8.506440e+05	2.930084e+05	2.951546e+05	2.283974e+04	2.396392e+05	1.821947e+05	5.433809e+04	3106.426507	2016.147899
std	0.402677	3.453545e+06	1.264989e+06	1.204120e+06	1.074641e+05	9.862424e+05	7.461785e+05	2.439660e+05	17692.894652	0.939938
min	0.440000	8.456000e+01	0.000000e+00	0.000000e+00	0.000000e+00	0.000000e+00	0.000000e+00	0.000000e+00	0.000000	2015.000000
25%	1.100000	1.083858e+04	8.540700e+02	3.008780e+03	0.000000e+00	5.088640e+03	2.849420e+03	1.274700e+02	0.000000	2015.000000
50%	1.370000	1.073768e+05	8.645300e+03	2.906102e+04	1.849900e+02	3.974383e+04	2.636282e+04	2.647710e+03	0.000000	2016.000000
75%	1.660000	4.329623e+05	1.110202e+05	1.502069e+05	6.243420e+03	1.107834e+05	8.333767e+04	2.202925e+04	132.500000	2017.000000
max	3.250000	6.250565e+07	2.274362e+07	2.047057e+07	2.546439e+06	1.937313e+07	1.338459e+07	5.719097e+06	551693.650000	2018.000000

In [12]:

df['year'].unique()

Out[12]:

array([2015, 2016, 2017, 2018])

In [13]:

df['region'].nunique()

Out[13]:

In [14]:

df['region'].unique()

Out[14]:

array(['Albany', 'Atlanta', 'BaltimoreWashington', 'Boise', 'Boston',
       'BuffaloRochester', 'California', 'Charlotte', 'Chicago',
       'CincinnatiDayton', 'Columbus', 'DallasFtWorth', 'Denver',
       'Detroit', 'GrandRapids', 'GreatLakes', 'HarrisburgScranton',
       'HartfordSpringfield', 'Houston', 'Indianapolis', 'Jacksonville',
       'LasVegas', 'LosAngeles', 'Louisville', 'MiamiFtLauderdale',
       'Midsouth', 'Nashville', 'NewOrleansMobile', 'NewYork',
       'Northeast', 'NorthernNewEngland', 'Orlando', 'Philadelphia',
       'PhoenixTucson', 'Pittsburgh', 'Plains', 'Portland',
       'RaleighGreensboro', 'RichmondNorfolk', 'Roanoke', 'Sacramento',
       'SanDiego', 'SanFrancisco', 'Seattle', 'SouthCarolina',
       'SouthCentral', 'Southeast', 'Spokane', 'StLouis', 'Syracuse',
       'Tampa', 'TotalUS', 'West', 'WestTexNewMexico'], dtype=object)

In [ ]:

필요없는 맨 처음 컬럼을 제거하시오¶

In [ ]:

데이터의 날짜가 뒤죽박죽 입니다. 날짜로 정렬하시오.¶

In [18]:

df.sort_values('Date', inplace= True)

In [19]:

df

Out[19]:

	Date	AveragePrice	Total Volume	4046	4225	4770	Total Bags	Small Bags	Large Bags	XLarge Bags	type	year	region
51	2015-01-04	1.75	27365.89	9307.34	3844.81	615.28	13598.46	13061.10	537.36	0.00	organic	2015	Southeast
51	2015-01-04	1.49	17723.17	1189.35	15628.27	0.00	905.55	905.55	0.00	0.00	organic	2015	Chicago
51	2015-01-04	1.68	2896.72	161.68	206.96	0.00	2528.08	2528.08	0.00	0.00	organic	2015	HarrisburgScranton
51	2015-01-04	1.52	54956.80	3013.04	35456.88	1561.70	14925.18	11264.80	3660.38	0.00	conventional	2015	Pittsburgh
51	2015-01-04	1.64	1505.12	1.27	1129.50	0.00	374.35	186.67	187.68	0.00	organic	2015	Boise
...	...	...	...	...	...	...	...	...	...	...	...	...	...
0	2018-03-25	1.36	908202.13	142681.06	463136.28	174975.75	127409.04	103579.41	22467.04	1362.59	conventional	2018	Chicago
0	2018-03-25	0.70	9010588.32	3999735.71	966589.50	30130.82	4014132.29	3398569.92	546409.74	69152.63	conventional	2018	SouthCentral
0	2018-03-25	1.42	163496.70	29253.30	5080.04	0.00	129163.36	109052.26	20111.10	0.00	organic	2018	SouthCentral
0	2018-03-25	1.70	190257.38	29644.09	70982.10	0.00	89631.19	89424.11	207.08	0.00	organic	2018	California
0	2018-03-25	1.34	1774776.77	63905.98	908653.71	843.45	801373.63	774634.09	23833.93	2905.61	conventional	2018	NewYork

18249 rows × 13 columns

날짜별로 가격이 어떻게 변하는지 간단하게 확인하시오. (plot 이용)¶

In [24]:

df_date = df.groupby('Date')['AveragePrice'].mean()

In [25]:

df_date

Out[25]:

Date
2015-01-04    1.301296
2015-01-11    1.370648
2015-01-18    1.391111
2015-01-25    1.397130
2015-02-01    1.247037
                ...   
2018-02-25    1.359630
2018-03-04    1.350185
2018-03-11    1.335093
2018-03-18    1.313704
2018-03-25    1.346852
Name: AveragePrice, Length: 169, dtype: float64

In [27]:

df_date.plot()
plt.show()

In [ ]:

# 주기성이 있다.

In [ ]:

'region' 별로 데이터 몇개인지 시각화 하시오.¶

In [28]:

df

Out[28]:

	Date	AveragePrice	Total Volume	4046	4225	4770	Total Bags	Small Bags	Large Bags	XLarge Bags	type	year	region
51	2015-01-04	1.75	27365.89	9307.34	3844.81	615.28	13598.46	13061.10	537.36	0.00	organic	2015	Southeast
51	2015-01-04	1.49	17723.17	1189.35	15628.27	0.00	905.55	905.55	0.00	0.00	organic	2015	Chicago
51	2015-01-04	1.68	2896.72	161.68	206.96	0.00	2528.08	2528.08	0.00	0.00	organic	2015	HarrisburgScranton
51	2015-01-04	1.52	54956.80	3013.04	35456.88	1561.70	14925.18	11264.80	3660.38	0.00	conventional	2015	Pittsburgh
51	2015-01-04	1.64	1505.12	1.27	1129.50	0.00	374.35	186.67	187.68	0.00	organic	2015	Boise
...	...	...	...	...	...	...	...	...	...	...	...	...	...
0	2018-03-25	1.36	908202.13	142681.06	463136.28	174975.75	127409.04	103579.41	22467.04	1362.59	conventional	2018	Chicago
0	2018-03-25	0.70	9010588.32	3999735.71	966589.50	30130.82	4014132.29	3398569.92	546409.74	69152.63	conventional	2018	SouthCentral
0	2018-03-25	1.42	163496.70	29253.30	5080.04	0.00	129163.36	109052.26	20111.10	0.00	organic	2018	SouthCentral
0	2018-03-25	1.70	190257.38	29644.09	70982.10	0.00	89631.19	89424.11	207.08	0.00	organic	2018	California
0	2018-03-25	1.34	1774776.77	63905.98	908653.71	843.45	801373.63	774634.09	23833.93	2905.61	conventional	2018	NewYork

18249 rows × 13 columns

In [29]:

df['region'].value_counts()

Out[29]:

Southeast              338
NewOrleansMobile       338
SanDiego               338
BaltimoreWashington    338
Roanoke                338
RichmondNorfolk        338
Northeast              338
SouthCentral           338
GreatLakes             338
Louisville             338
Seattle                338
CincinnatiDayton       338
NewYork                338
Indianapolis           338
Chicago                338
Jacksonville           338
Columbus               338
Detroit                338
Philadelphia           338
PhoenixTucson          338
Nashville              338
Portland               338
HartfordSpringfield    338
Tampa                  338
Orlando                338
West                   338
Denver                 338
GrandRapids            338
NorthernNewEngland     338
BuffaloRochester       338
HarrisburgScranton     338
Pittsburgh             338
Boise                  338
LosAngeles             338
LasVegas               338
Atlanta                338
DallasFtWorth          338
MiamiFtLauderdale      338
Plains                 338
StLouis                338
Syracuse               338
Midsouth               338
Sacramento             338
Boston                 338
Charlotte              338
Spokane                338
Albany                 338
Houston                338
SouthCarolina          338
SanFrancisco           338
TotalUS                338
RaleighGreensboro      338
California             338
WestTexNewMexico       335
Name: region, dtype: int64

In [30]:

import seaborn as sb

In [36]:

plt.figure(figsize= (6,11) )
sb.countplot(data=df, y = 'region')
plt.show()

In [ ]:

년도('year')별로 데이터가 몇건인지 확인하시오.¶

In [52]:

plt.figure(figsize= (10,5))
sb.countplot(data = df , y = 'year')
plt.show()

In [53]:

df.tail()

Out[53]:

Date	AveragePrice	Total Volume	4046	4225	4770	Total Bags	Small Bags	Large Bags	XLarge Bags	type	year	region
2018-03-25	1.36	908202.13	142681.06	463136.28	174975.75	127409.04	103579.41	22467.04	1362.59	conventional	2018	Chicago
2018-03-25	0.70	9010588.32	3999735.71	966589.50	30130.82	4014132.29	3398569.92	546409.74	69152.63	conventional	2018	SouthCentral
2018-03-25	1.42	163496.70	29253.30	5080.04	0.00	129163.36	109052.26	20111.10	0.00	organic	2018	SouthCentral
2018-03-25	1.70	190257.38	29644.09	70982.10	0.00	89631.19	89424.11	207.08	0.00	organic	2018	California
2018-03-25	1.34	1774776.77	63905.98	908653.71	843.45	801373.63	774634.09	23833.93	2905.61	conventional	2018	NewYork

In [ ]:

프로펫 분석을 위해, 두개의 컬럼만 가져오시오. ('Date', 'AveragePrice')¶

In [42]:

# 프로펫은 날짜와 날짜별로 예측할것 두가지가 필수로 있어야함.
avocado_prophet_df = df[ [ 'Date', 'AveragePrice' ] ]

In [43]:

avocado_prophet_df

Out[43]:

	Date	AveragePrice
51	2015-01-04	1.75
51	2015-01-04	1.49
51	2015-01-04	1.68
51	2015-01-04	1.52
51	2015-01-04	1.64
...	...	...
0	2018-03-25	1.36
0	2018-03-25	0.70
0	2018-03-25	1.42
0	2018-03-25	1.70
0	2018-03-25	1.34

18249 rows × 2 columns

STEP 3: Prophet 을 이용한 예측 수행¶

ds 와 y 로 컬럼명을 셋팅하시오.¶

In [45]:

# 프로펫은 컬럼명이 통일되어있어야함.
avocado_prophet_df.columns = ['ds', 'y']

In [46]:

avocado_prophet_df

Out[46]:

	ds	y
51	2015-01-04	1.75
51	2015-01-04	1.49
51	2015-01-04	1.68
51	2015-01-04	1.52
51	2015-01-04	1.64
...	...	...
0	2018-03-25	1.36
0	2018-03-25	0.70
0	2018-03-25	1.42
0	2018-03-25	1.70
0	2018-03-25	1.34

18249 rows × 2 columns

프로펫 예측 하시오.¶

In [54]:

# 1. 라이브러리를 변수로 만들고,
prophet = Prophet()

In [55]:

# 2. 데이터로, 학습 시킨다.
prophet.fit(avocado_prophet_df)

INFO:prophet:Disabling weekly seasonality. Run prophet with weekly_seasonality=True to override this.
INFO:prophet:Disabling daily seasonality. Run prophet with daily_seasonality=True to override this.

Out[55]:

<prophet.forecaster.Prophet at 0x7f5c6478ee50>

In [57]:

# 365일치를 예측하시오.
# 3. 예측하고자 하는 기간을 정해서, 비어있는 데이터 프레임을 만든다!.
future = prophet.make_future_dataframe(periods=365, freq='D') # periods=365, freq='D' == 365일  // periods=365, freq='W' == 365주

In [58]:

future

Out[58]:

	ds
0	2015-01-04
1	2015-01-11
2	2015-01-18
3	2015-01-25
4	2015-02-01
...	...
529	2019-03-21
530	2019-03-22
531	2019-03-23
532	2019-03-24
533	2019-03-25

534 rows × 1 columns

In [59]:

# 4. 이제, 미래 날짜까지 만들어져 있으니,
#    위의 future 데이터프레임을 이용해서, 예측할 수 있다.
forecast = prophet.predict(future)

In [60]:

forecast # yhat == 예측값

Out[60]:

	ds	trend	yhat_lower	yhat_upper	trend_lower	trend_upper	additive_terms	additive_terms_lower	additive_terms_upper	yearly	yearly_lower	yearly_upper	multiplicative_terms	multiplicative_terms_lower	multiplicative_terms_upper	yhat
0	2015-01-04	1.498608	0.879976	1.874224	1.498608	1.498608	-0.113270	-0.113270	-0.113270	-0.113270	-0.113270	-0.113270	0.0	0.0	0.0	1.385337
1	2015-01-11	1.493471	0.906704	1.877692	1.493471	1.493471	-0.104849	-0.104849	-0.104849	-0.104849	-0.104849	-0.104849	0.0	0.0	0.0	1.388621
2	2015-01-18	1.488334	0.889276	1.831219	1.488334	1.488334	-0.104524	-0.104524	-0.104524	-0.104524	-0.104524	-0.104524	0.0	0.0	0.0	1.383810
3	2015-01-25	1.483198	0.842019	1.809866	1.483198	1.483198	-0.123469	-0.123469	-0.123469	-0.123469	-0.123469	-0.123469	0.0	0.0	0.0	1.359729
4	2015-02-01	1.478061	0.821152	1.803670	1.478061	1.478061	-0.151828	-0.151828	-0.151828	-0.151828	-0.151828	-0.151828	0.0	0.0	0.0	1.326232
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
529	2019-03-21	1.161728	0.542878	1.613888	0.980910	1.336850	-0.086221	-0.086221	-0.086221	-0.086221	-0.086221	-0.086221	0.0	0.0	0.0	1.075508
530	2019-03-22	1.160997	0.562707	1.580983	0.979392	1.336800	-0.084549	-0.084549	-0.084549	-0.084549	-0.084549	-0.084549	0.0	0.0	0.0	1.076448
531	2019-03-23	1.160266	0.517173	1.591213	0.977661	1.336861	-0.082604	-0.082604	-0.082604	-0.082604	-0.082604	-0.082604	0.0	0.0	0.0	1.077662
532	2019-03-24	1.159535	0.540519	1.601641	0.975953	1.337079	-0.080406	-0.080406	-0.080406	-0.080406	-0.080406	-0.080406	0.0	0.0	0.0	1.079129
533	2019-03-25	1.158804	0.576940	1.572218	0.974279	1.337226	-0.077982	-0.077982	-0.077982	-0.077982	-0.077982	-0.077982	0.0	0.0	0.0	1.080822

534 rows × 16 columns

In [ ]:

# 차트로 확인하시오.

In [63]:

prophet.plot(forecast) # 차트그리는 첫번째 방법
plt.savefig('chart1.jpg') # 이미지로 저장하라 // 두개로 나타내는 버그를 없앤다. 

In [65]:

prophet.plot_components(forecast) # 차트그리는 두번째 방법
plt.savefig('chart2.jpg')

PART 2 : region 이 west 인 아보카도의 가격을 예측하시오.¶

In [88]:

sorted (df['region'].unique())

Out[88]:

['Albany',
 'Atlanta',
 'BaltimoreWashington',
 'Boise',
 'Boston',
 'BuffaloRochester',
 'California',
 'Charlotte',
 'Chicago',
 'CincinnatiDayton',
 'Columbus',
 'DallasFtWorth',
 'Denver',
 'Detroit',
 'GrandRapids',
 'GreatLakes',
 'HarrisburgScranton',
 'HartfordSpringfield',
 'Houston',
 'Indianapolis',
 'Jacksonville',
 'LasVegas',
 'LosAngeles',
 'Louisville',
 'MiamiFtLauderdale',
 'Midsouth',
 'Nashville',
 'NewOrleansMobile',
 'NewYork',
 'Northeast',
 'NorthernNewEngland',
 'Orlando',
 'Philadelphia',
 'PhoenixTucson',
 'Pittsburgh',
 'Plains',
 'Portland',
 'RaleighGreensboro',
 'RichmondNorfolk',
 'Roanoke',
 'Sacramento',
 'SanDiego',
 'SanFrancisco',
 'Seattle',
 'SouthCarolina',
 'SouthCentral',
 'Southeast',
 'Spokane',
 'StLouis',
 'Syracuse',
 'Tampa',
 'TotalUS',
 'West',
 'WestTexNewMexico']

In [74]:

df_west = df.loc[ df['region'] == 'West', ] 

In [77]:

df_west = df_west[ [ 'Date', 'AveragePrice']] # 안가져와도 컬럼명만 바꾸면 알아서 예측해주긴한다.

In [78]:

df_west

Out[78]:

	Date	AveragePrice
51	2015-01-04	1.40
51	2015-01-04	0.89
50	2015-01-11	1.39
50	2015-01-11	0.95
49	2015-01-18	0.96
...	...	...
2	2018-03-11	1.00
1	2018-03-18	1.73
1	2018-03-18	0.99
0	2018-03-25	0.93
0	2018-03-25	1.60

338 rows × 2 columns

In [79]:

df_west.columns = ['ds','y'] # rename 해도됨.

In [80]:

df_west

Out[80]:

	ds	y
51	2015-01-04	1.40
51	2015-01-04	0.89
50	2015-01-11	1.39
50	2015-01-11	0.95
49	2015-01-18	0.96
...	...	...
2	2018-03-11	1.00
1	2018-03-18	1.73
1	2018-03-18	0.99
0	2018-03-25	0.93
0	2018-03-25	1.60

338 rows × 2 columns

In [81]:

prophet2 = Prophet()

In [82]:

prophet2.fit( df_west )

INFO:prophet:Disabling weekly seasonality. Run prophet with weekly_seasonality=True to override this.
INFO:prophet:Disabling daily seasonality. Run prophet with daily_seasonality=True to override this.

Out[82]:

<prophet.forecaster.Prophet at 0x7f5c6486bf10>

In [89]:

future2 = prophet2.make_future_dataframe(periods=365, freq='D')

In [90]:

future2

Out[90]:

	ds
0	2015-01-04
1	2015-01-11
2	2015-01-18
3	2015-01-25
4	2015-02-01
...	...
529	2019-03-21
530	2019-03-22
531	2019-03-23
532	2019-03-24
533	2019-03-25

534 rows × 1 columns

In [91]:

forecast2 = prophet2.predict(future2)

In [92]:

forecast2

Out[92]:

	ds	trend	yhat_lower	yhat_upper	trend_lower	trend_upper	additive_terms	additive_terms_lower	additive_terms_upper	yearly	yearly_lower	yearly_upper	multiplicative_terms	multiplicative_terms_lower	multiplicative_terms_upper	yhat
0	2015-01-04	1.287090	0.639895	1.520475	1.287090	1.287090	-0.188138	-0.188138	-0.188138	-0.188138	-0.188138	-0.188138	0.0	0.0	0.0	1.098952
1	2015-01-11	1.284842	0.701152	1.559720	1.284842	1.284842	-0.172230	-0.172230	-0.172230	-0.172230	-0.172230	-0.172230	0.0	0.0	0.0	1.112611
2	2015-01-18	1.282593	0.663041	1.538315	1.282593	1.282593	-0.163686	-0.163686	-0.163686	-0.163686	-0.163686	-0.163686	0.0	0.0	0.0	1.118907
3	2015-01-25	1.280345	0.706518	1.547059	1.280345	1.280345	-0.175002	-0.175002	-0.175002	-0.175002	-0.175002	-0.175002	0.0	0.0	0.0	1.105342
4	2015-02-01	1.278096	0.669138	1.481156	1.278096	1.278096	-0.196058	-0.196058	-0.196058	-0.196058	-0.196058	-0.196058	0.0	0.0	0.0	1.082038
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
529	2019-03-21	1.725246	1.091646	1.927508	1.685382	1.759265	-0.207836	-0.207836	-0.207836	-0.207836	-0.207836	-0.207836	0.0	0.0	0.0	1.517409
530	2019-03-22	1.725805	1.092608	1.951162	1.685880	1.759976	-0.202810	-0.202810	-0.202810	-0.202810	-0.202810	-0.202810	0.0	0.0	0.0	1.522995
531	2019-03-23	1.726364	1.069243	1.962666	1.686323	1.760687	-0.197214	-0.197214	-0.197214	-0.197214	-0.197214	-0.197214	0.0	0.0	0.0	1.529150
532	2019-03-24	1.726924	1.108538	1.944338	1.686720	1.761397	-0.191153	-0.191153	-0.191153	-0.191153	-0.191153	-0.191153	0.0	0.0	0.0	1.535771
533	2019-03-25	1.727483	1.106256	1.981002	1.687123	1.762108	-0.184741	-0.184741	-0.184741	-0.184741	-0.184741	-0.184741	0.0	0.0	0.0	1.542743

534 rows × 16 columns

In [96]:

prophet2.plot(forecast2)
plt.savefig('chart3.png')

In [ ]:

In [97]:

prophet2.plot_components(forecast2) # 차트그리는 두번째 방법
plt.savefig('chart4.png')

In [ ]:

Pandas 시카고 범죄율을 예측 Prophet, error_bad_lines, to_datetime(format), resample 함수의 사용법과, 이 함수를 사용하기 위해 인덱스를 설정하는 방법 (0)	2023.01.03
Pandas datetime,datetime64,데이터프레임 날짜 일괄처리 (0)	2022.11.30
Pandas Tip[1] 문자열 컬럼의 슬라이싱. str (0)	2022.11.30
Pandas concat(), merge() 여러 데이터 프레임을 하나로 합치는 방법 (0)	2022.11.25
Pandas 데이터프레임 오름차순, 내림차순 정렬 .Sort_values() ,sort_index() (0)	2022.11.25

블로그안함

Pandas Prophet 라이브러리를 이용한 Time Series 데이터 예측 방법

AVOCADO 가격 예측 (Facebook Prophet )¶

STEP #0: 데이터셋¶

STEP #1: 데이터 준비¶

Prophet 라이브러리¶

STEP #2: EDA(Exploratory Data Analysis) : 탐색적 데이터 분석¶

필요없는 맨 처음 컬럼을 제거하시오¶

데이터의 날짜가 뒤죽박죽 입니다. 날짜로 정렬하시오.¶

날짜별로 가격이 어떻게 변하는지 간단하게 확인하시오. (plot 이용)¶

'region' 별로 데이터 몇개인지 시각화 하시오.¶

년도('year')별로 데이터가 몇건인지 확인하시오.¶

프로펫 분석을 위해, 두개의 컬럼만 가져오시오. ('Date', 'AveragePrice')¶

STEP 3: Prophet 을 이용한 예측 수행¶

ds 와 y 로 컬럼명을 셋팅하시오.¶

프로펫 예측 하시오.¶

PART 2 : region 이 west 인 아보카도의 가격을 예측하시오.¶

'DataScience > Pandas' 카테고리의 다른 글

티스토리툴바

Pandas Prophet 라이브러리를 이용한 Time Series 데이터 예측 방법

AVOCADO 가격 예측 (Facebook Prophet )¶

STEP #0: 데이터셋¶

STEP #1: 데이터 준비¶

Prophet 라이브러리¶

STEP #2: EDA(Exploratory Data Analysis) : 탐색적 데이터 분석¶

필요없는 맨 처음 컬럼을 제거하시오¶

데이터의 날짜가 뒤죽박죽 입니다. 날짜로 정렬하시오.¶

날짜별로 가격이 어떻게 변하는지 간단하게 확인하시오. (plot 이용)¶

'region' 별로 데이터 몇개인지 시각화 하시오.¶

년도('year')별로 데이터가 몇건인지 확인하시오.¶

프로펫 분석을 위해, 두개의 컬럼만 가져오시오. ('Date', 'AveragePrice')¶

STEP 3: Prophet 을 이용한 예측 수행¶

ds 와 y 로 컬럼명을 셋팅하시오.¶

프로펫 예측 하시오.¶

PART 2 : region 이 west 인 아보카도의 가격을 예측하시오.¶

'DataScience > Pandas' 카테고리의 다른 글

관련글

티스토리툴바