import random
import pandas as pd

# バッティング結果の確率を設定
batting_results = {
    "homerun": 0.05652173913043478, # 本塁打
    "triple": 0.008695652173913044, # 三塁打
    "double": 0.0782608695652174,  # 二塁打
    "single": 0.1565217391304348,  # 安打
    "strike": 0.1826086956521739, # 三振
    "walk": 0.10869565217391304, # 四球
    "out":  0.40869565217391307 # その他のアウト
}

def simulate_batting():
    rand = random.random()
    cumulative_probability = 0.0
    for result, probability in batting_results.items():
        cumulative_probability += probability
        if rand < cumulative_probability:
            return result
    return "out"

base = []
for i in range(1000):
    tmp = {k:0 for k in batting_results.keys()}
    n = 650
    for i in range(n):
        s = simulate_batting()
        if s in tmp:
            tmp[s]+=1
        else:
            tmp[s]=1
    base.append(tmp)

df = pd.DataFrame(base)
df["hits"] = df["homerun"] + df["triple"] + df["double"] + df["single"]
df.describe()

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

# homerunの平均値と標準偏差を計算
mean = df['homerun'].mean()
std = df['homerun'].std()
print(f"{int(mean-std)}, {int(mean)}, {int(mean+std)}")

# ヒストグラムを表示
df['homerun'].hist(bins=20, alpha=0.7)

# 平均値、標準偏差±1のラインを描画
plt.axvline(x=mean, color='red', linestyle='--', label='mean')
plt.axvline(x=mean+std, color='green', linestyle='--', label='std+1')
plt.axvline(x=mean-std, color='blue', linestyle='--', label='std-1')

# 凡例を表示
plt.legend()
plt.show()

# hitsの平均値と標準偏差を計算
mean = df['hits'].mean()
std = df['hits'].std()
print(f"{int(mean-std)}, {int(mean)}, {int(mean+std)}")

# ヒストグラムを表示
df['hits'].hist(bins=20, alpha=0.7)

# 平均値、標準偏差±1のラインを描画
plt.axvline(x=mean, color='red', linestyle='--', label='mean')
plt.axvline(x=mean+std, color='green', linestyle='--', label='std+1')
plt.axvline(x=mean-std, color='blue', linestyle='--', label='std-1')

# 凡例を表示
plt.legend()
plt.show()

タグ ：

#プログラミング

#Python

#大谷選手

#メジャーリーグ

#予測

import matplotlib.pyplot as plt
import pandas as pd
%matplotlib inline

df = pd.read_csv("https://oku.edu.mie-u.ac.jp/~okumura/python/data/COVID-19.csv",
                 index_col='Date', parse_dates=['Date'])

df.iloc[0:,0:2].plot(figsize=(16,9))
plt.show()

df['diff_Global Confirmed'] = df['Global Confirmed'].pct_change()
df['diff_Global Deaths'] = df['Global Deaths'].pct_change()
plt.figure(figsize=(16,9))
plt.bar(df.index,df['diff_Global Confirmed'],alpha=0.3,label="Confirmed")
plt.bar(df.index,df['diff_Global Deaths'],alpha=0.3,label="Deaths")
plt.legend()
plt.show()

df['Confirmed2']=df['Global Confirmed'] - df['China Confirmed']
df['Deaths2']=df['Global Deaths'] - df['China Deaths']
df.iloc[0:,6:8].plot(figsize=(16,9))
plt.show()

df['diff_Confirmed2'] = df['Confirmed2'].pct_change()
df['diff_Deaths2'] = df['Deaths2'].pct_change()
plt.figure(figsize=(16,9))
plt.bar(df.index,df['diff_Confirmed2'],alpha=0.3,label="Confirmed")
plt.bar(df.index,df['diff_Deaths2'],alpha=0.3,label="Deaths")
plt.legend()
plt.show()

df['Death Rate'] = df['Deaths2'] / df['Confirmed2']
plt.figure(figsize=(16,9))
plt.bar(df.index,df['Death Rate'],alpha=0.8,label="Death Rate")
plt.legend()
plt.show()

from fbprophet import Prophet
import warnings

warnings.simplefilter('ignore',)
model_df = df[['Confirmed2']]
model_df = model_df.reset_index()
model_df = model_df.rename(columns={'Date':'ds','Confirmed2':'y'})
model_df.head()

model = Prophet()
model.fit(model_df)

future_df = model.make_future_dataframe(600)
forecast_df = model.predict(future_df)
model.plot(forecast_df)

タグ ：

#Python

#プログラミング

#予測

#新型コロナウィルス