Add function “get_crypto_visuals” to timeseries.py

datasist/timeseries.py

@@ -6,8 +6,11 @@
 import pandas as pd
 import numpy as np
 import matplotlib.pyplot as plt
+import random
+import yfinance as yf
+from plotly import graph_objs as go
 import seaborn as sns
-from IPython.display import display
+from IPython.display import display, Image
 from .structdata import get_cat_feats, get_num_feats, get_date_cols
 
 
@@ -226,7 +229,7 @@ def get_period_of_day(date_col=None):
         raise ValueError("date_cols: Expect a date columns, got 'None'")
 
 
-    if date_col.dtype != np.int:
+    if date_col.dtype != np.int64:
 
         date_col_hr = pd.to_datetime(date_col).dt.hour
         return date_col_hr.map(_map_hours)
@@ -388,4 +391,168 @@ def _map_hours(x):
     elif x in [13,14,15,16]:
         return 'afternoon'
     else:
-        return 'evening'
+        return 'evening'
+
+
+def get_crypto_visuals(coin, period="5d", interval="15m", MA=False, days=[7,25,99], boll=False, boll_sma=25, save_fig=False, img_format="png"):
+
+
+
+
+  '''
+
+  Utilizing Plotly and yfinance to make a Stock Market Coin-USD timeseries plot using Candlesticks
+
+
+
+  Parameters
+    -----------------
+
+      coin: str
+
+                - Symbol of Crypto coin i.e 'BTC', 'ETH', 'DOGE', 'BNB', 'SXP'
+                - Must be currently listed on https://finance.yahoo.com/cryptocurrencies/
+
+
+      period: str, Default "5d"
+
+                - Answers the question - How far back from realtime should data be sourced?
+                - Valid periods: '1d','5d','1mo','3mo','6mo','1y','2y','5y','10y','ytd','max'
+
+
+      interval: str, Default "15m"
+
+                - References intra-period intervals of data
+                - Valid intervals: '1m','2m','5m','15m','30m','60m','90m','1h','1d','5d','1wk','1mo','3mo'
+                - Intraday data cannot extend last 60 days
+
+
+      MA: bool, Default False
+
+                - References the plotting of the Moving Average (MA) data
+
+
+
+      days: list, tuple, 1D array, Pandas Series, Set, Default = [7, 25, 99]
+
+                - Must be an iterable containing integer values only
+                - The integers represent, day subsets for the Moving Average data
+
+
+      boll: bool, Default False
+
+                - References the plotting of the bollinger band
+                - Important: • When both MA and boll are True, MA takes preference
+                             • The above algorithm is in place to avoid a rowdy chart
+
+      boll_sma:   Integer, Default 25             
+
+                - Indicates the SMA value used as a reference point for plotting the bollinger band
+
+
+      save_fig:   bool, Default False
+
+                - Saves the plot to the current working directory
+
+
+      img_format: str, Default "png"
+                - The desired image format:'png', 'jpg' or 'jpeg', 'webp', 'svg', and 'pdf'
+
+
+
+
+  Returns
+    -----------------
+       None
+
+
+  Examples of valid use
+    -----------------
+    >>> get_crypto_visuals("BTC")
+    >>> get_crypto_visuals("BTC", MA=True)
+    >>> get_crypto_visuals("BTC", period="10d", interval="5m", MA=True, days=[10,30,100], save_fig=True, img_format='jpeg')
+    >>> get_crypto_visuals("BTC", period="10d", interval="5m", MA=False, days=[10,30,100], boll=True)
+
+  '''
+
+  #Error Handling
+  if coin is None: raise ValueError("coin: Symbol of Crypto coin i.e 'BTC', 'ETH', 'DOGE', 'BNB', 'SXP'\nMust be currently listed on https://finance.yahoo.com/cryptocurrencies/")
+  if not isinstance(period, str): raise TypeError('period: This parameter should be a string')
+  if not isinstance(interval, str): raise TypeError('interval: This parameter should be a string')
+  if not isinstance(img_format, str): raise TypeError('img_format: This parameter should be a string')
+  if not isinstance(coin, str): raise TypeError('coin: This parameter should be a string')
+  if not isinstance(MA, bool): raise TypeError('MA: This parameter should be boolean (True or False)')
+  if not isinstance(boll, bool): raise TypeError('boll: This parameter should be boolean (True or False)')
+  if not isinstance(save_fig, bool): raise TypeError('save_fig: This parameter should be boolean (True or False)')
+  if not isinstance(boll_sma, int): raise TypeError('boll_sma: This parameter should be an Integer')
+  if (not hasattr(days, '__iter__')) or isinstance(days, str): raise TypeError("days: This parameter should be iterable, strings aren't also allowed")
+
+
+
+
+
+  coin = coin.upper().strip()
+  data = yf.download(tickers=f'{coin}-USD',period = period, interval = interval)
+  header = f"{coin}-USD Price Analysis"
+
+
+  #Generate Moving Average
+  if MA==True:
+    for i in days: data[f'MA{i}'] = data['Close'].rolling(i).mean()
+
+
+  #Generate Bollinger
+  def get_bollinger_band(prices, rate=boll_sma):
+    sma = prices.rolling(rate).mean()
+    std = prices.rolling(rate).std()
+    bollinger_up = sma + std * 2 
+    bollinger_down = sma - std * 2
+    return sma, bollinger_up, bollinger_down  
+
+  if boll:
+    if not (MA or (boll and MA)): sma, u,d = get_bollinger_band(data['Close'])
+
+  #declare figure
+  fig = go.Figure()
+
+
+  #Candlestick
+  fig.add_trace(go.Candlestick(x=data.index,
+                               open=data['Open'],
+                               high=data['High'],
+                               low=data['Low'],
+                               close=data['Close'], 
+                               name = 'Market data'))
+  #Randomly pick from preferred colors
+  col_lst = ['pink','darkgray','orange','darkblue','darkcyan','darkgoldenrod','darkgray','darkgrey','darkkhaki','darkmagenta','darkorange','darkorchid','darksalmon','darkslateblue','darkslategray','darkslategrey','darkturquoise','darkviolet','deeppink']
+  try: col_sam = random.sample(col_lst, len(days))
+  except ValueError: col_sam = np.random.choice(col_lst, len(days))
+
+
+  #Add Moving average on the chart
+  co = 0
+  for col in data.columns:
+    if col.startswith('MA'):
+      fig.add_trace(go.Scatter(x=data.index, y= data[col],line={'color':col_sam[co],'width':1.5}, name = col))
+      co+=1
+
+  #Add Bollinger on the chart
+  if boll:
+    if not (MA or (boll and MA)):
+       fig.add_trace(go.Scatter(x=data.index, y= u,line={'color':'darkmagenta','width':1.5}, name = "BOLL (Up)"))
+       fig.add_trace(go.Scatter(x=data.index, y=sma, line={'color':'orange','width':1.5}, name="BOLL (Mid)"))
+       fig.add_trace(go.Scatter(x=data.index, y= d,line={'color':'darkblue','width':1.5}, name = "BOLL (Down)"))
+
+
+  #Updating axis and graph
+  fig.update_xaxes(title=f'Datetime', rangeslider_visible =True)
+
+  fig.update_yaxes(title='USD ($)')
+  fig.update_layout({"title": {"text": header}})
+
+  #Show
+  fig.show()
+
+
+  #Save
+  if save_fig: fig.write_image("fig_timeplot_{}-usd.{}".format(coin.lower(), img_format.lower()), engine='kaleido')

requirements.txt

@@ -20,6 +20,7 @@ jupyter==1.0.0
 jupyter-client==6.1.3
 jupyter-console==6.1.0
 jupyter-core==4.6.3
+kaleido==0.2.1
 kiwisolver==1.2.0
 MarkupSafe==1.1.1
 matplotlib==3.2.2
@@ -36,16 +37,17 @@ pandocfilters==1.4.2
 parso==0.7.0
 pexpect==4.8.0
 pickleshare==0.7.5
+plotly==0.5.0
 pluggy==0.13.1
-ppscore==0.0.3
+ppscore==1.2.0
 prometheus-client==0.8.0
 prompt-toolkit==3.0.5
 ptyprocess==0.6.0
 py==1.10.0
 Pygments==2.7.4
 pyparsing==2.4.7
 pyrsistent==0.16.0
-pytest==5.4.3
+pytest==6.2.4
 python-dateutil==2.8.1
 pytz==2020.1
 pyzmq==19.0.1
@@ -66,4 +68,5 @@ traitlets==4.3.3
 wcwidth==0.2.5
 webencodings==0.5.1
 widgetsnbextension==3.5.1
+yfinance==0.1.63
 zipp==3.1.0