Tutorial 1
This is a beginner tutorial, focusing on getting the data downloaded, loading the cache and getting it into a window.
$ pip install -U finplot
Collecting finplot
Using cached finplot-1.7-py3-none-any.whl (33 kB)
Requirement already satisfied: pandas in d:\prog\python39\lib\site-packages (from finplot) (1.2.3)
Requirement already satisfied: pyqtgraph>=0.11.1 in d:\prog\python39\lib\site-packages (from finplot) (0.12.0)
Requirement already satisfied: PyQt5 in d:\prog\python39\lib\site-packages (from finplot) (5.15.2)
Requirement already satisfied: numpy>=1.17.0 in d:\prog\python39\lib\site-packages (from pyqtgraph>=0.11.1->finplot) (1.20.2)
Requirement already satisfied: python-dateutil>=2.7.3 in d:\prog\python39\lib\site-packages (from pandas->finplot) (2.8.1)
Requirement already satisfied: pytz>=2017.3 in d:\prog\python39\lib\site-packages (from pandas->finplot) (2020.4)
Requirement already satisfied: six>=1.5 in d:\prog\python39\lib\site-packages (from python-dateutil>=2.7.3->pandas->finplot) (1.15.0)
Requirement already satisfied: PyQt5-sip<13,>=12.8 in d:\prog\python39\lib\site-packages (from PyQt5->finplot) (12.8.1)
Installing collected packages: finplot
Successfully installed finplot-1.7Ok, so we start by caching some data on your local computer. This has nothing to do with finplot per se, but many seem to struggle with this. Create a script called download.py or something like that.
import yfinance as yf
df = yf.download('VWAGY', '2014-05-01') # Wolksvagen
df.to_csv('cache.csv')Save and let's see what we've got:
$ python download.py
[*********************100%***********************] 1 of 1 completed
$ head cache.csv
Date,Open,High,Low,Close,Adj Close,Volume
2014-04-30,26.635000228881836,26.729999542236328,26.489999771118164,26.729999542236328,25.87717056274414,24800
2014-05-01,26.683500289916992,26.683500289916992,26.59000015258789,26.594999313354492,25.746477127075195,13800
2014-05-02,26.125,26.229999542236328,26.059999465942383,26.1200008392334,25.28663444519043,38600
2014-05-05,25.78499984741211,26.29400062561035,25.56999969482422,26.264999389648438,25.4270076751709,429000
2014-05-06,25.8799991607666,26.03499984741211,25.760000228881836,25.975000381469727,25.14626121520996,100200
2014-05-07,25.915000915527344,26.065000534057617,25.75749969482422,25.822500228881836,24.998626708984375,75800
2014-05-08,26.104999542236328,26.2450008392334,25.969999313354492,26.0,25.170461654663086,93000
2014-05-09,25.940000534057617,25.94499969482422,25.264999389648438,25.450000762939453,24.638011932373047,101200
2014-05-12,25.94499969482422,25.9950008392334,25.399999618530273,25.454999923706055,24.642850875854492,42800So far so good. Note that the date format is ISO-8601.
Create a file called plot-downloaded.py:
import pandas as pd
import finplot as fplt
df = pd.read_csv('cache.csv')
print(df)This gets us the flat, raw data. The print at the end looks like this:
$ python plot-downloaded.py
Date Open High Low Close Adj Close Volume
0 2014-04-30 26.635000 26.730000 26.490000 26.730000 25.877171 24800
1 2014-05-01 26.683500 26.683500 26.590000 26.594999 25.746477 13800
2 2014-05-02 26.125000 26.230000 26.059999 26.120001 25.286634 38600
3 2014-05-05 25.785000 26.294001 25.570000 26.264999 25.427008 429000
4 2014-05-06 25.879999 26.035000 25.760000 25.975000 25.146261 100200
... ... ... ... ... ... ... ...
1774 2021-05-17 32.290001 32.738998 32.290001 32.694000 32.694000 318100
1775 2021-05-18 32.599998 32.750000 32.360001 32.450001 32.450001 301800
1776 2021-05-19 32.220001 32.664001 31.879999 32.650002 32.650002 282700
1777 2021-05-20 32.993000 33.130001 32.840000 33.055000 33.055000 347200
1778 2021-05-21 33.150002 33.599998 32.950001 33.549999 33.549999 541800
[1779 rows x 7 columns]Here, pandas has lost a piece of information. The Date column used to be the index, but now we've instead gotten a zero-based index, and Date has turned into a column. This is how Pandas works by default.
So for the moment we are particularly interested in the Date column, and how to turn it into a date index once again. Add another print statement at the end to show some more info on the Date column:
print(df['Date'])Run again:
$ python plot-downloaded.py
...
0 2014-04-30
1 2014-05-01
2 2014-05-02
3 2014-05-05
4 2014-05-06
...
1774 2021-05-17
1775 2021-05-18
1776 2021-05-19
1777 2021-05-20
1778 2021-05-21
Name: Date, Length: 1779, dtype: objectNotice two things. The type of the column is object (in this case strings). And also, we would like to use date and time as index. To convert strings to dates and use as index:
df['Date'] = pd.to_datetime(df['Date'])
df = df.set_index('Date')
print(df)The result:
$ python plot-downloaded.py
...
Open High Low Close Adj Close Volume
Date
2014-04-30 26.635000 26.730000 26.490000 26.730000 25.877171 24800
2014-05-01 26.683500 26.683500 26.590000 26.594999 25.746477 13800
2014-05-02 26.125000 26.230000 26.059999 26.120001 25.286634 38600
2014-05-05 25.785000 26.294001 25.570000 26.264999 25.427008 429000
2014-05-06 25.879999 26.035000 25.760000 25.975000 25.146261 100200
... ... ... ... ... ... ...
2021-05-17 32.290001 32.738998 32.290001 32.694000 32.694000 318100
2021-05-18 32.599998 32.750000 32.360001 32.450001 32.450001 301800
2021-05-19 32.220001 32.664001 31.879999 32.650002 32.650002 282700
2021-05-20 32.993000 33.130001 32.840000 33.055000 33.055000 347200
2021-05-21 33.150002 33.599998 32.950001 33.549999 33.549999 541800
[1779 rows x 6 columns]Notice that the Date label belongs to the index now. This is good news.
All we need to do now is create the plot and show the window:
fplt.candlestick_ochl(df[['Open', 'Close', 'High', 'Low']])
fplt.show()The _ochl part is short for open, close, high, low — i.e. the order of columns to the candlestick plot function.
