日常問題解決彙總

日常操作時遇到各種報錯或新想法，所遇問題以及搜索到的解決方案參考來源大彙總（持續更新）

Python數據科學庫

1. URLError: <urlopen error [Errno 61] Connection refused>

# 原代碼
import seaborn as sns
planets = sns.load_dataset('planets')

解決：從github下載數據集，用pandas調用

• mwaskom/seaborn-data - GitHub

import pandas as pd
planets = pd.read_csv('./seaborn-data-master/planets.csv')

2. Keyerror: ‘Date’

• Keyerror ‘Date’ when using pandas datareader - Stackoverflow

I received the ‘Keyerror ‘Date’ when using pandas datareader’ error and found two errors in my script that fixed the issue:

1. The name of the entity was incorrect, for example using ‘APPL’ instead of ‘AAPL’.
2. There was no data for the date parameters I was using.

# 解決代碼
# Use 'BTC-USD' stock/security value
import pandas as pd
import pandas_datareader.data as web
import datetime as dt

start = dt.datetime(2017, 1, 1)
end = dt.datetime(2019, 11, 30)

df = web.DataReader('BTC-USD', 'yahoo', start, end)
df.to_csv('BTC.csv')
print(df.head())

• 如不確定數據集起始時間也可以設置時間，如下

yhoo = data.DataReader('BTC-USD', data_source='yahoo')
yhoo.head()

yhoo.tail()

• 可以看出數據集起始時間是2015年至2020年

3. NotImplementedError: data_source=None is not implemented

# 原代碼
from pandas_datareader impoart data
goog = data.DataReader('GOOG', start='2004', end='2016')
goog.head()

• Issue when running DataReader - pandas-datareader issue - GitHub

As of v0.7.0 Google finance and Morningstar have been immediately deprecated due to large changes in their API and no stable replacement.

---

終端：

1. pip install 下載庫時中途報錯，切換網絡重試

• 詭異的錯誤：OpenSSL.SSL.Error: [(‘SSL routines’, ‘ssl3_get_record’, ‘decryption failed or bad record mac’)

---

Markdown：

1. 正文序號下繼續分級，多空幾個空格可以實現

• 研究了一個markdown做多級編號

---