def candlestick(self, ticker, start = None, end = None):
data = self[ticker][start:end]
quotes = zip(date2num(data.index.astype(datetime.date)), data.Open, data.High, data.Low, data.Close)
fig, ax = plt.subplots()
candlestick_ohlc(ax, quotes)
ax.xaxis_date()
fig.autofmt_xdate()
return (fig, ax)
def plot_candlestick(df):
"""
Plots a candlestick chart of the time series
:param df:
:return:
"""
# creating a new DataFrame based on the adjusted_close price resampled with a 10 day window
tickers_ohlc = df['adj_close'].resample('10D').ohlc()
# creating a new DataFrame based on the adjusted volume resampled with a 10 day window
tickers_volume = df['adj_volume'].resample('10D').sum()
# resetting the index of the DataFrame
tickers_ohlc = tickers_ohlc.reset_index()
# converting the date column to mdates
tickers_ohlc['date'] = tickers_ohlc['date'].map(mdates.date2num)
# creating a new figure
fig = plt.figure()
# creating a subplot with a 6x1 grid and starts at (0,0)
ax1 = plt.subplot2grid((6, 1), (0, 0), rowspan=5, colspan=1)
# creating a subplot with a 6x1 grid and starts at (5,0)
ax2 = plt.subplot2grid((6, 1), (5, 0), rowspan=1, colspan=1, sharex=ax1)
# converts the axis from raw mdate numbers to dates
ax1.xaxis_date()
# plotting the candlestick graph
candlestick_ohlc(ax1, tickers_ohlc.values, width=2, colorup='g')
# plotting the volume bar chart
ax2.fill_between(tickers_volume.index.map(mdates.date2num),
tickers_volume.values, 0)
fig.tight_layout()
def plot_candlestic_with_extrema(dataframe, peak_indices, trough_indices,
stock_name):
fig, ax = plt.subplots(figsize=(16, 7))
fig.subplots_adjust(bottom=0.2)
quotes = zip(mdates.date2num(dataframe.index.to_pydatetime()),
dataframe[u'Open'], dataframe[u'High'], dataframe[u'Low'],
dataframe[u'Close'])
low_price_seq = dataframe['Low'].values
high_price_seq = dataframe['High'].values
candlestick_ohlc(ax,
quotes,
width=0.75,
colorup='g',
colordown='red',
alpha=0.6)
plt.plot(dataframe.index[peak_indices],
pd.Series(high_price_seq, index=dataframe.index)[peak_indices],
"v",
color='black')
plt.plot(dataframe.index[trough_indices],
pd.Series(low_price_seq, index=dataframe.index)[trough_indices],
"^",
color='blue')
ax.xaxis_date()
ax.legend([stock_name], loc='upper right', shadow=True, fancybox=True)
ax.autoscale_view()
plt.setp(plt.gca().get_xticklabels(),
rotation=45,
horizontalalignment='right')
plt.rc('axes', grid=True)
plt.rc('grid', color='0.75', linestyle='-', linewidth=0.5)
plt.show()
def changeCandle():
plt.clf()
if candleType == "1H":
ax1 = plt.subplot2grid((6, 1), (0, 0), rowspan=5, colspan=1)
ax2 = plt.subplot2grid((6, 1), (5, 0), rowspan=1, colspan=1, sharex=ax1)
plt.setp(ax1.get_xticklabels(), visible=False)
if showDates == False:
plt.setp(ax2.get_xticklabels(), visible=False)
df_ohlc = df_segment[["Close"]].resample(candleType).sum()
df_volume = df_segment["Volume To"].resample(candleType).sum()
df_ohlc.reset_index(inplace=True)
df_ohlc["Date"] = df_ohlc["Date"].map(mdates.date2num)
ax1.xaxis_date() # show mdates as readable normal date
plt.plt(ax1)
ax2.fill_between(df_volume.index.map(mdates.date2num), df_volume.values, 0, facecolors=volumeColor)
fig.canvas.draw()
else:
ax1 = plt.subplot2grid((6, 1), (0, 0), rowspan=5, colspan=1)
ax2 = plt.subplot2grid((6, 1), (5, 0), rowspan=1, colspan=1, sharex=ax1)
plt.setp(ax1.get_xticklabels(), visible=False)
if showDates == False:
plt.setp(ax2.get_xticklabels(), visible=False)
df_ohlc = df_segment["Close"].resample(candleType).ohlc()
df_volume = df_segment["Volume To"].resample(candleType).sum()
df_ohlc.reset_index(inplace=True)
df_ohlc["Date"] = df_ohlc["Date"].map(mdates.date2num)
ax1.xaxis_date() # show mdates as readable normal date
candlestick_ohlc(ax1, df_ohlc.values, width=candleWidth, colorup=lightColor, colordown=darkColor)
ax2.fill_between(df_volume.index.map(mdates.date2num), df_volume.values, 0, facecolors=volumeColor)
fig.canvas.draw()
def plot_candlestick(self, data_in):
"""
Add volume: https://stackoverflow.com/questions/13128647/matplotlib-finance-volume-overlay
:param data_in: list of list [[bar.date, bar.open, bar.high, bar.low, bar.close, bar.volume], [...], ...]
:return:
"""
# Loop over all candles
# data_in = [1,1]
ohlc_data = []
self.axes.cla()
for i in range(0, len(data_in)):
date = data_in[i][BAR_DICT['Date']]
conv_date = datetime.datetime.strptime(date, '%Y%m%d')
# data_in[i][1] = data_in[i][1].astype(float)
# data_in[i][2] = data_in[i][2].astype(float)
# data_in[i][3] = data_in[i][3].astype(float)
# data_in[i][4] = data_in[i][4].astype(float)
# data_in[i][5] = data_in[i][5].astype(float)
ohlc = []
ohlc.append(date2num(conv_date))
ohlc.append(float(data_in[i][1]))
ohlc.append(float(data_in[i][2]))
ohlc.append(float(data_in[i][3]))
ohlc.append(float(data_in[i][4]))
ohlc.append(float(data_in[i][5]))
ohlc_data.append(ohlc)
candlestick_ohlc(self.axes,
ohlc_data,
width=0.4,
colorup='#77d879',
colordown='#db3f3f')
self.axes.xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m-%d'))
self.axes.xaxis.set_major_locator(mticker.MaxNLocator(10))
self.fig.autofmt_xdate()
self.axes.grid(True)
self.draw()
def plot_custom_plot(self, cs_data, no_time=False):
self.clear_sub_plots()
candle_list = []
dates = [f['datetime'] / 1000.0 for f in cs_data]
if not no_time:
for candle in cs_data:
candle_list.append([
float(mdates.epoch2num(candle['datetime'] / 1000.0)),
float(candle['open']),
float(candle['high']),
float(candle['low']),
float(candle['close']),
float(candle['volume'])
])
else:
for x in range(len(cs_data)):
candle_list.append([
x,
float(cs_data[x]['open']),
float(cs_data[x]['high']),
float(cs_data[x]['low']),
float(cs_data[x]['close']),
float(cs_data[x]['volume'])
])
# Changing the labels for the x-axis
# self.axis_list[3].xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m-%d'))
# dates_formatted = [mdates.epoch2num(candle['datetime'] / 1000.) for candle in cs_data['candles']]
# print('Dates formatted %s' % dates_formatted)
self.axis_list[1].get_xaxis().set_visible(False)
self.axis_list[2].get_xaxis().set_visible(False)
candlestick_ohlc(self.axis_list[2],
candle_list,
colorup='green',
colordown='red',
width=.001)
def pattern(df, rule, signal):
if signal == 'BearishHarami' or signal == 'BullishHarami':
last1, last2 = 8, 9
elif signal == 'MorningStar' or signal == 'EveningStar':
last1, last2 = 7, 8
target = df
fontsize = 12
plt.rcParams['xtick.labelsize'] = fontsize
plt.rcParams['ytick.labelsize'] = fontsize
plt.rcParams['axes.titlesize'] = fontsize
fig = plt.figure(figsize=(24, 8))
ax = plt.subplot2grid((1, 1), (0, 0))
ax.set_xticks(range(10))
ax.set_xticklabels(target.index)
y = target['close'].iloc[0:last1].values.reshape(-1, 1)
x = np.array(range(1, last2)).reshape(-1, 1)
model = LinearRegression()
model.fit(x, y)
y_pred = model.predict(x)
ax.plot(y_pred, label='Trend')
arr = np.c_[range(target.shape[0]),
target[['open', 'high', 'low', 'close']].values]
mpf.candlestick_ohlc(ax,
arr,
width=0.4,
alpha=1,
colordown='#53c156',
colorup='#ff1717')
locs, labels = plt.xticks()
plt.setp(labels, rotation=45)
plt.grid()
ax.legend(loc='best', prop={'size': fontsize})
title_name = signal + '_' + rule
ax.set_title(title_name)
fig.subplots_adjust(bottom=0.25)
name = signal + '_' + rule
plt.savefig(name)
plt.show()
def prepare(self, chart_data):
# Initialize the canvas and prepare to draw 4 charts
self.fig, self.axes = plt.subplots(nrows=4,
ncols=1,
facecolor='w',
sharex=True)
for ax in self.axes:
# Disable scientific notations that are hard to read
ax.get_xaxis().get_major_formatter().set_scientific(False)
ax.get_yaxis().get_major_formatter().set_scientific(False)
# Chart 1. Daily Chart
self.axes[0].set_ylabel('Env.') # y 축 레이블 표시
# Volume visualization
x = np.arange(len(chart_data))
volume = np.array(chart_data)[:, -1].tolist()
self.axes[0].bar(chart_data["date"], volume, color='b', alpha=0.3)
# ohlc stands for open, high, low and close, it is a two-dimensional array in this order
ax = self.axes[0].twinx()
ax.set_title("Daily Chart")
ohlc = np.hstack((x.reshape(-1, 1), np.array(chart_data)[:, 1:-1]))
# Bar chart output to self.axes [0]
# Positive chart is in red, negative chart is in blue
candlestick_ohlc(ax, ohlc, colorup='r', colordown='b')
def graph_main_old(tag_df, start='2017-09-01', code=None):
data = get_stock_data2(tag_df, start)
# 将数据导入数据库
#df_in_db(data, code)
with open('./dataSets/data.temp', 'wb') as f:
f.write(pickle.dumps([data, code]))
prices = data[['open', 'high', 'low', 'close']]
prices_lst = prices.values.tolist()
prices_lst.reverse()
prices = pd.DataFrame(prices_lst, columns=['open', 'high', 'low', 'close'])
dates = data['date_time'].apply(
lambda x: datetime.datetime.strftime(x, "%Y-%m-%d"))
candleData = np.column_stack([list(range(len(dates))), prices])
fig = plt.figure(figsize=(10, 6))
ax = fig.add_axes([0.1, 0.3, 0.8, 0.6])
ax.xaxis.set_major_locator(ticker.MultipleLocator(1))
#ax.set_xticklabels(dates, rotation=45)
ax.grid(True, linestyle='-.')
mpf.candlestick_ohlc(ax, candleData, width=0.5, colorup='r', colordown='b')
plt.savefig('./static/images/test.png')
#plt.show()
return labels_util(dates, candleData), code
def matplotRealTime(i):
ohlc = []
endTime = time.time()
dataCandle = login.get_candles('EURUSD', 60, 100, endTime)
dataIndex = 1
if len(dataCandle) > 0:
for data in dataCandle:
candleData = dataIndex, data['open'], data['max'], data[
'min'], data['close'], data['volume']
ohlc.append(candleData)
dataIndex += 1
ax1.clear()
ax1.grid(True)
plt.xlabel('Date')
plt.ylabel('Price')
plt.title("IQ OPTION")
candlestick_ohlc(ax1,
ohlc,
width=0.4,
colorup='#77d879',
colordown='#db3f3f')
def drop_k(symbol):
#获取数据，并将日期类型转换为matplotlib日期
data = nasdaq_mysql.getData(symbol)
new_data = []
for row in data:
row = list(row)
row[0] = mdata.date2num(datetime.strptime(str(row[0]), "%Y-%m-%d"))
new_data.append(row)
#创建子图,得到一个二元组，图表对象和坐标轴对象
fig, ax = plt.subplots(figsize=(10, 5))
#调整子图参数，bottom是设置图表到底部的距离
fig.subplots_adjust(bottom=0.2)
plt.ylabel('股价（元）')
plt.xlabel('时间')
plt.title('股票代码：{0}近三个月历史记录'.format(symbol))
mpl.candlestick_ohlc(ax, new_data, width=1.0, colorup='r', colordown='g')
#设置坐标轴属性
ax.xaxis_date() #日期类型
plt.xticks(rotation=45) #显示刻度值旋转45度
plt.savefig('股票K线图.png')
plt.show()
def get_plot_candlesticks_dataframe(data_frame, ax):
candlestick_ohlc(ax,
data_frame[PriceStrings.STR_PRICE_OPEN.value],
data_frame[PriceStrings.STR_PRICE_HIGH.value],
data_frame[PriceStrings.STR_PRICE_LOW.value],
data_frame[PriceStrings.STR_PRICE_CLOSE.value],
width=0.8,
colorup='#008000',
colordown='#FF0000',
alpha=1)
# xdate = [datetime.datetime.fromtimestamp(i) for i in dataframe[PriceStrings.STR_PRICE_TIME.value]]
xdate = data_frame[PriceStrings.STR_PRICE_TIME.value].values
ax.xaxis.set_major_locator(ticker.MaxNLocator(6))
def get_date(x, _):
try:
return xdate[int(x)]
except IndexError:
return ''
ax.xaxis.set_major_formatter(ticker.FuncFormatter(get_date))
def draw_candlestick():
index = np.arange(5)
columns = np.array(['date', 'open', 'high', 'low', 'close'])
datas = [
(0, 12, 14, 12, 12.5),
(1, 12, 14, 12, 12.5),
(2, 12, 14, 12, 12.5),
(3, 12, 14, 12, 12.5),
(4, 12, 14, 12, 12.5),
]
df = pd.DataFrame(datas, index=index, columns=columns)
df = df.as_matrix()
print(df)
fig, ax = plt.subplots()
# ax.xaxis_date()
plt.xticks(rotation=45)
mpf.candlestick_ohlc(ax, df, width=1.0, colorup='r', colordown='green', alpha=1) ##设置利用mpf画股票K线图
date_tickers = ['20190101', '20190102', '20190103', '20190104', '20190105']
ax.set_xticks(range(len(date_tickers)))
ax.set_xticklabels(date_tickers)
ax.set_xticklabels(date_tickers)
plt.show()
return False
def gett(): #button(确定)
#股票代码
w.start()
plt.rcParams['font.sans-serif'] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
dailyQuota = w.wsd(aa, "open,high,low,close", statime, endtime,
"Fill=Previous")
print(dailyQuota.Times)
print(len(dailyQuota.Data))
tupleQuota = []
for i in range(len(dailyQuota.Data[0])):
tupleQuota.append((dailyQuota.Times[i].toordinal(),
dailyQuota.Data[0][i], dailyQuota.Data[1][i],
dailyQuota.Data[2][i], dailyQuota.Data[3][i]))
print(tupleQuota)
mondayFormatter = DateFormatter('%Y-%m-%d')
fig, ax = plt.subplots()
ax.xaxis.set_major_formatter(mondayFormatter)
candlestick_ohlc(ax, tupleQuota, width=0.4, colorup='r', colordown='g')
plt.xticks(rotation=30)
plt.title(aa)
plt.show()
def plt_chart_atr(chart, path=os.path.join(os.getcwd(), "tmp")):
df_tmp_full = chart.sort_values(by=["date"], ascending=True)
df_tmp = df_tmp_full.ix[:, [
'open_price', 'close_price', 'high_price', 'low_price'
]]
move_avrg_lst = [5, 10, 20, 40, 80]
fig = plt.figure()
ax = plt.subplot()
xdate = [x for x in df_tmp_full.date] # 日付
ochl = np.vstack((date2num(xdate), df_tmp.values.T)).T
# mpf.candlestick_ochl(ax, ochl, width=0.7, colorup='g', colordown='r')
candlestick_ohlc(ax1, ochl, width=0.7, colorup='g', colordown='r')
ax.grid() # グリッド表示
ax.set_xlim(df_tmp_full.iloc[0].date, df_tmp_full.iloc[-1].date) # x軸の範囲
fig.autofmt_xdate() # x軸のオートフォーマット
plt.plot(df_tmp_full["date"], df_tmp_full["atr"], label="atr")
plt.legend()
plt.savefig(os.path.join(path, chart["stock_id"].values[0] +
'_chart.png')) #画像保存
def plot_candlestick(df, ax=None, fmt="%Y-%m-%d"):
if ax is None:
fig, ax = plt.subplots()
idx_name = df.index.name
print(idx_name)
dat = df.reset_index()[[idx_name, "Open", "High", "Low", "Close"]]
dat[df.index.name] = dat[df.index.name].map(mdates.date2num)
ax.xaxis_date()
ax.xaxis.set_major_formatter(mdates.DateFormatter(fmt))
plt.xticks(rotation=45)
_ = candlestick_ohlc(ax, dat.values, width=.6, colorup='g', alpha=1)
ax.set_xlabel(idx_name)
ax.set_ylabel("OHLC")
return ax
def draw_k(hist_data):
data_list = []
for dates, row in hist_data.iterrows():
date_time = datetime.datetime.strptime(dates, '%Y-%m-%d')
t = date2num(date_time)
open, high, close, low = row[:4]
datas = (t, open, high, low, close
) # tushare里的数据顺序为open,high,close,low注意
data_list.append(datas)
fig, ax = plt.subplots(figsize=(16, 10))
fig.subplots_adjust(bottom=0.2)
mpf.candlestick_ohlc(ax,
data_list,
width=1.5,
colorup='r',
colordown='green')
plt.grid()
ax.xaxis_date()
ax.autoscale_view()
plt.setp(plt.gca().get_xticklabels(), rotation=30)
plt.show()
def plot_stock(self, k_data):
from mpl_finance import candlestick_ohlc
self.priceMax = k_data.high.values.max()
self.dateMin = k_data.time.values.min()
self.dateMax = k_data.time.values.max()
candlestick_ohlc(self.price_ax,
k_data.values,
width=1.0,
colorup='r',
colordown='g')
self.price_ax.plot(k_data.time,
k_data['uprice'],
'b',
label="无穷成本均线",
linewidth=1)
self.price_ax.set_ylabel("prices")
self.price_ax.yaxis.label.set_color('k')
self.price_ax.set_xlim(self.dateMin, self.dateMax)
self.price_ax.set_ylim(self.priceMin, self.priceMax)
self.price_ax.xaxis.set_major_locator(mticker.MultipleLocator(250))
self.price_ax.xaxis.set_major_formatter(
mticker.FuncFormatter(self.format_date))
self.price_ax.grid(True, color='k', linestyle='--')
def plot():
plt.style.use('ggplot')
data = pd.read_csv("NVDA.csv")
data = data.loc[:, ['Date', 'Open', 'High', 'Low', 'Close']]
data['Date'] = pd.to_datetime(data['Date'])
data['Date'] = data['Date'].apply(mpl_dates.date2num)
data = data.astype(float)
# Creating Subplots
fig, ax = plt.subplots()
candlestick_ohlc(ax, data.values, width=0.6, colorup='green', colordown='red', alpha=0.8)
# Setting labels & titles
ax.set_xlabel('Date')
ax.set_ylabel('Price')
fig.suptitle('NVIDIA STOCK PRICE CANDLESTICK CHART')
# Formatting Date
date_format = mpl_dates.DateFormatter('%d-%m-%Y')
ax.xaxis.set_major_formatter(date_format)
fig.autofmt_xdate()
fig.tight_layout()
plt.show()
def plot(self, start_date, end_date, index_code):
df, index_data = self.get_data(start_date, end_date, index_code)
date_tickers = index_data.date.tolist()
def _format_date(x, pos=None):
if x < 0 or x > len(date_tickers) - 1: return ''
return date_tickers[int(x)]
info = self.compute_stock_score(df)
candlestick_ohlc(self.price_ax,
index_data.values,
width=1.0,
colorup='r',
colordown='g')
self.ratio_ax.plot(info['date'],
info['rate'],
'r',
label="超跌系数",
linewidth=1)
self.price_ax.xaxis.set_major_locator(mticker.MultipleLocator(20))
self.price_ax.xaxis.set_major_formatter(
mticker.FuncFormatter(_format_date))
plt.show()
def prepare(self, chart_data):
# 캔버스를 초기화하고 4개의 차트를 그릴 준비
self.fig, self.axes = plt.subplots(nrows=4, ncols=1, facecolor='w', sharex='all')
for ax in self.axes:
# 보기 어려운 과학적 표기 비활성화
ax.get_xaxis().get_majer_formatter().set_scientific(False)
ax.get_yaxis().get_major_formatter().set_scientific(False)
# 차트 1. 일봉 차트
self.axes[0].set_ylabel('Env.') # y 축 레이블 표시
# 거래량 가시화
x = np.arange(len(chart_data))
volume = np.array(chart_data)[:, -1].tolist()
self.axes[0].bar(x, volume, color='b', alpha=0.3)
# ohlc 란 open, high, low, close 의 약자로 이 순서로 구성된 2차원 배열
ax = self.axes[0].twinx()
ohlc = np.hstack((x.reshape(-1, 1), np.array(chart_data)[:, 1:-1]))
# self.axes[0]에 봉 차트 출력
# 양봉은 빨간색으로, 음봉은 파란색으로 표시
candlestick_ohlc(ax, ohlc, colorup='r', colordown='b')
def showStock(stock):
df = pd.read_csv(stock + '.csv', parse_dates=True, index_col=0)
df['100ma'] = df['Adj Close'].rolling(window=100, min_periods=0).mean()
df_ohlc = df['Adj Close'].resample('10D').ohlc()
df_volume = df['Volume'].resample('10D').sum()
df_ohlc.reset_index(inplace=True)
df_ohlc['Date'] = df_ohlc['Date'].map(mdates.date2num)
ax1 = plt.subplot2grid((6, 1), (0, 0), rowspan=5, colspan=1)
ax2 = plt.subplot2grid((6, 1), (5, 0), rowspan=1, colspan=1, sharex=ax1)
ax1.xaxis_date()
candlestick_ohlc(ax1, df_ohlc.values, width=2, colorup='g')
ax2.fill_between(df_volume.index.map(mdates.date2num), df_volume.values, 0)
#ax1.plot(df.index, df['Adj Close'])
#ax1.plot(df.index, df['100ma'])
#ax2.bar(df.index, df['Volume'])
plt.show()
def candle_plot_time_series(self,
scrip_code,
df,
recent=0,
show=True,
save=True):
Validator.validate_attribute(recent, int, True)
count, steps, dfmt = self.__plot_params(df, recent)
figure, axis = plt.subplots(figsize=(16, 8))
# create a copy of the DataFrame to operate on
dohlc = df.tail(count).copy()
# take the index back to column
dohlc.reset_index(inplace=True)
# drop the closed price column, adj.closed price will be considered
dohlc.drop('Close', axis=1, inplace=True)
# convert the datetime format to short string format
dohlc.Date = dohlc.Date.apply(DateTime.dateformatter_short)
# get hold of dates for title naming
str_dates = dohlc.Date
# convert the string dates to pandas TimeStamp values
dohlc.Date = pd.to_datetime(dohlc.Date, format='%d-%m-%Y')
# convert the pandas TimeStamp values to matplotlib float values
dohlc.Date = dohlc.Date.apply(mdates.date2num)
candlestick_ohlc(ax=axis,
quotes=dohlc.values,
width=0.6,
colorup='green',
colordown='red',
alpha=0.7)
axis.xaxis.set_major_formatter(mdates.DateFormatter(dfmt))
axis.set_xticks(dohlc.Date[::-steps])
plt.setp(plt.gca().get_xticklabels(), rotation=90)
start_date = str_dates.iloc[0]
end_date = str_dates.iloc[-1]
plt.title(scrip_code + ' (' + start_date + ' to ' + end_date + ')')
plt.tight_layout()
self.__show_and_save(plt, show, save, 'candlesticks.png')
def plt_chart_w_rsi(self, path=os.path.join(os.getcwd(), "chart")):
real_id_lst = self.df["stock_id"].drop_duplicates()
for id_str in real_id_lst:
df_tmp_full = self.df[self.df.stock_id == id_str].sort_values(
by=["date"], ascending=True)
df_tmp = df_tmp_full.ix[:, [
'open_price', 'high_price', 'low_price', 'close_price'
]]
fig = plt.figure()
ax1 = plt.subplot(211)
ax2 = plt.subplot(212)
xdate = [x for x in df_tmp_full.date] # 日付
ochl = np.vstack((date2num(xdate), df_tmp.values.T)).T
# mpf.candlestick_ochl(ax1, ochl, width=0.7, colorup='g', colordown='r')
candlestick_ohlc(ax1, ochl, width=0.7, colorup='g', colordown='r')
ax1.grid() # グリッド表示
ax1.set_xlim(df_tmp_full.iloc[0].date,
df_tmp_full.iloc[-1].date) # x軸の範囲
fig.autofmt_xdate() # x軸のオートフォーマット
close_price_lst = df_tmp_full["close_price"].values.flatten()
ax2.grid() # グリッド表示
ax2.set_xlim(df_tmp_full["date"].values[0],
df_tmp_full["date"].values[-1]) # x軸の範囲
fig.autofmt_xdate() # x軸のオートフォーマット
plt.plot(df_tmp_full["date"].values,
df_tmp_full["rsi"].values,
label="rsi_14")
plt.legend()
if not os.path.isdir(path):
os.makedirs(path)
plt.savefig(os.path.join(path,
str(id_str) + '_chart_w_rsi.png')) #画像保存
def candlestick(comp_name, filter_data, cmp_tick, from_date_s, to_date_s):
#sort data into open, high, low, close format
filter_prices = filter_data.loc[:, ["Open", "High", "Low", "Close"]]
#Reset index to get dates as a seperate column
filter_prices = filter_prices.reset_index()
#conver date to numbers
filter_prices['Date'] = filter_prices['Date'].apply(mpl_dates.date2num)
#convert to float
filter_prices = filter_prices.astype(float)
#Create subplot
graph, plot = plt.subplots(2, sharex=True, figsize=(15, 6))
#create candle stick graph on plot
candlestick_ohlc(plot[0],
filter_prices.values,
width=0.3,
colorup="green",
colordown="red")
#convert number back to dates
dates_from_num = mpl_dates.DateFormatter('%d-%m-%Y')
graph.suptitle(
"Candlestick graph v/s Volume for {} -> {} from {} - {}".format(
cmp_tick, comp_name, from_date_s, to_date_s))
plot[0].set_title("Candlestick Graph")
plot[0].xaxis.set_major_formatter(dates_from_num)
plot[0].set_ylabel("Stock Price")
plot[1].set_title("Volume")
plot[1].bar(filter_data.index, filter_data.Volume)
plot[1].set_xlabel("Time")
plot[1].set_ylabel("Volume")
plt.show()
def draw_candlestick_on_ax(ax,
stock_code,
key_type,
need_update,
key_count=73):
key_quotes, date_quotes, low_quotes, high_quotes = klinedata.load_klinedata(
stock_code, key_type, need_update, key_count)
mpf.candlestick_ohlc(ax,
key_quotes,
colordown='#53c156',
colorup='#ff1717',
width=0.2,
alpha=1) #colordown='#F5F5F5', colorup='#DCDCDC'
# #https://matplotlib.org/examples/pylab_examples/date_index_formatter.html
class MyFormatter(ticker.Formatter):
def __init__(self, fmt='%Y-%m-%d %H:%M'):
self.fmt = fmt
def __call__(self, x, pos=0): #x就是x轴的刻度数值，但是是浮点数
if x < 0 or x >= len(date_quotes):
return ''
# slice seconds
return date_quotes[int(x)][:-3] if date_quotes[int(
x)][-3:-2] == ':' else date_quotes[int(x)]
# set xaxix format
formatter = MyFormatter(date_quotes)
ax.xaxis.set_major_formatter(formatter)
ax.xaxis.set_major_locator(ticker.MultipleLocator(len(date_quotes) //
4)) # 根据数据量控制间隔
for label in ax.get_xticklabels():
label.set_horizontalalignment('center')
add_guides(stock_code, key_type, low_quotes, high_quotes)
def candleVolume(seriesData,candletitle='a',bartitle='b'):
Date=[date2num(date) for date in seriesData.index]
seriesData.index=list(range(len(Date)))
seriesData['Date']=Date
listData=zip(seriesData.Date,seriesData.Open,seriesData.High,seriesData.Low,
seriesData.Close)
ax1 = plt.subplot(211)
ax2 = plt.subplot(212)
for ax in ax1,ax2:
mondays = WeekdayLocator(MONDAY)
weekFormatter = DateFormatter('%m/%d/%Y')
ax.xaxis.set_major_locator(mondays)
ax.xaxis.set_minor_locator(DayLocator())
ax.xaxis.set_major_formatter(weekFormatter)
ax.grid(True)
ax1.set_ylim(seriesData.Low.min()-2,seriesData.High.max()+2)
ax1.set_ylabel('蜡烛图及收盘价线')
candlestick_ohlc(ax1,listData, width=0.7,colorup='r',colordown='g')
plt.setp(plt.gca().get_xticklabels(),\
rotation=45,horizontalalignment='center')
ax1.autoscale_view()
ax1.set_title(candletitle)
ax1.plot(seriesData.Date,seriesData.Close,\
color='black',label='收盘价')
ax1.legend(loc='best')
ax2.set_ylabel('成交量')
ax2.set_ylim(0,seriesData.Volume.max()*3)
ax2.bar(np.array(Date)[np.array(seriesData.Close>=seriesData.Open)]
,height=seriesData.iloc[:,4][np.array(seriesData.Close>=seriesData.Open)]
,color='r',align='center')
ax2.bar(np.array(Date)[np.array(seriesData.Close<seriesData.Open)]
,height=seriesData.iloc[:,4][np.array(seriesData.Close<seriesData.Open)]
,color='g',align='center')
ax2.set_title(bartitle)
return(plt.show())
def getcandlestick(stock):
data = get_history(symbol=stock,
start=date.today() - relativedelta(months=+3),
end=date.today())
data = data.reset_index()
data = data[['Date', 'Open', 'High', 'Low', 'Close']]
#visualization
plt.style.use('ggplot')
ohlc = data
ohlc['Date'] = pd.to_datetime(ohlc['Date'])
ohlc['Date'] = ohlc['Date'].apply(mpl_dates.date2num)
ohlc = ohlc.astype(float)
fig, ax = plt.subplots()
candlestick_ohlc(ax,
ohlc.values,
width=0.6,
colorup='green',
colordown='red',
alpha=0.8)
# Setting labels & titles
ax.set_xlabel('Date')
ax.set_ylabel('Price')
fig.suptitle('Daily Candlestick Chart of ' + stock)
# Formatting Date
date_format = mpl_dates.DateFormatter('%d-%m-%Y')
ax.xaxis.set_major_formatter(date_format)
fig.autofmt_xdate()
fig.tight_layout()
#mean
#ohlc['SMA5'] = ohlc["Close"].rolling(5).mean()
#ax.plot(ohlc['Date'], ohlc['SMA5'], color = 'green', label = 'SMA5')
plt.savefig('./media/candlestick.png', dpi=100, bbox_inches='tight')
def do_ml(ticker):
X, y, df = extract_featuresets(ticker, day_range)
X_train, X_test, y_train, y_test = cross_validation.train_test_split(
X, y, test_size=0.75)
# clf = neighbors.KNeighborsClassifier()
clf = VotingClassifier([('lsvc', svm.LinearSVC()),
('knn', neighbors.KNeighborsClassifier()),
('rfor', RandomForestClassifier())])
clf.fit(X_train, y_train)
confidence = clf.score(X_test, y_test)
print('Accuracy:', confidence)
predictions = clf.predict(X_test)
print('Predicted spread:', Counter(predictions))
df = pd.read_csv('stock_dfs/{}.csv'.format(ticker),
parse_dates=True,
index_col=0)
# df['100ma'] = df['Adj Close'].rolling(window=100, min_periods=0).mean()
df_ohlc = df['Adj Close'].resample('10D').ohlc()
df_volume = df['Volume'].resample('10D').sum()
df_ohlc.reset_index(inplace=True)
df_ohlc['Date'] = df_ohlc['Date'].map(mdates.date2num)
ax1 = plt.subplot2grid((6, 1), (0, 0), rowspan=5, colspan=1)
ax2 = plt.subplot2grid((6, 1), (5, 0), rowspan=1, colspan=1, sharex=ax1)
ax1.xaxis_date()
candlestick_ohlc(ax1,
df_ohlc.values,
width=2,
colorup='g',
colordown='r',
alpha=0.75)
ax2.fill_between(df_volume.index.map(mdates.date2num), df_volume.values, 0)
plt.show()
return confidence
def plot(self, n=50):
m = len(self.date)
if m == 0:
print('Markets.plot: No values saved, not posible to plot!')
return
elif m < n:
n = m
# Create figure
fig, ax = plt.subplots()
# Create subplot up
ax = plt.subplot(211)
# Candlestick plot
quotes = [
tuple([
self.date[i], self.open[i], self.high[i], self.low[i],
self.close[i]
]) for i in range(m - n, m)
]
candlestick_ohlc(ax, quotes, width=0.6)
# Value plot
plt.plot(self.date[m - n:m + 1], self.value[m - n:m + 1], 'y')
plt.ylabel('Value')
plt.title(self.name + ': CandleSticks-Value')
ax.xaxis.set_major_formatter(DateFormatter(''))
ax.grid(True)
# Create subplot down
ax = plt.subplot(212)
# Volume plot
plt.bar(self.date[m - n:m + 1], self.volume[m - n:m + 1], 0.6)
plt.ylabel('Volume')
plt.title(self.name + ': Volume')
ax.xaxis.set_major_formatter(DateFormatter('%b %d, %y'))
ax.grid(True)
# Format date, layout and print
fig.autofmt_xdate()
fig.tight_layout()
plt.show()
def stockPricePlot(code,start_date,date):
pro = ts.pro_api('5d9ec9dc6d71031a48a24b0e0f6c87e84fd2caf6bf15ac5845df7177') #token可以在新版tushare的网站上找到
history = pro.query('daily', ts_code = code, start_date = start_date, end_date = date)
history = history[::-1].reset_index()
print(history.head())
# 获取股票列表
stocklist = getlist()
name = []
print('stocklist:', len(stocklist))
if code in stocklist.keys():
name = stocklist[code]
print('name', name)
# 2. 数据操作
close = history[['close','open']]
# 将索引时间变成数据的一部分
close.dropna(inplace=True)
date_tickers = history['trade_date']
date_tickers = np.array(date_tickers.values)
# 变成时间戳
ohlc = history[['open','high','low','close']]
ohlc.reset_index()
ohlc['date'] = ohlc.index
# 将索引时间变成数据的一部分
ohlc = ohlc.dropna()
#print(ohlc)
# 3. 画图 k线图candle stick plot，散点图 scatter plot，
# 3.1 散点图 2行1列的图，散点图从0，0开始，占一行一列 x是tradedate，y是close，蓝色的
print(close[0:5],ohlc[0:5])
subplot1 = plt.subplot2grid((2,1),(0,0),rowspan=1,colspan=1)
# x轴变成date
#subplot1.xaxis_date()
subplot1.plot(date_tickers,close['open'],'black')
tick_spacing = 1
tick_spacing = 7
subplot1.xaxis.set_major_locator(ticker.MultipleLocator(tick_spacing))
plt.rcParams['font.sans-serif'] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
plt.title('\"'+code+'-'+str(name[0])+u'\"的股票历史记录 ')
# 3.2 k线图 为了在放大K线图的时候放大散点图，要共享x轴
subplot2 = plt.subplot2grid((2,1),(1,0),rowspan=1,colspan=1,sharex=subplot1)
a = mpf.candlestick_ohlc(ax=subplot2,quotes=ohlc[['date','open','high','low','close']].values,width=0.7,colorup='r',colordown='green', alpha=1)
plt.savefig('D:/software/python/project/stock/learn_envs/bishe_inter/result/history.png')
plt.show()
# (Year, month, day) tuples suffice as args for quotes_historical_yahoo
date1 = (2004, 2, 1)
date2 = (2004, 4, 12)
mondays = WeekdayLocator(MONDAY) # major ticks on the mondays
alldays = DayLocator() # minor ticks on the days
weekFormatter = DateFormatter('%b %d') # e.g., Jan 12
dayFormatter = DateFormatter('%d') # e.g., 12
quotes = quotes_historical_yahoo_ohlc('INTC', date1, date2)
if len(quotes) == 0:
raise SystemExit
fig, ax = plt.subplots()
fig.subplots_adjust(bottom=0.2)
ax.xaxis.set_major_locator(mondays)
ax.xaxis.set_minor_locator(alldays)
ax.xaxis.set_major_formatter(weekFormatter)
#ax.xaxis.set_minor_formatter(dayFormatter)
#plot_day_summary(ax, quotes, ticksize=3)
candlestick_ohlc(ax, quotes, width=0.6)
ax.xaxis_date()
ax.autoscale_view()
plt.setp(plt.gca().get_xticklabels(), rotation=45, horizontalalignment='right')
plt.show()
import datetime as dt
import matplotlib.pyplot as plt
from matplotlib import style
import pandas as pd
import pandas_datareader.data as web
from mpl_finance import candlestick_ohlc
import matplotlib.dates as mdates
pd.set_option('display.expand_frame_repr', False)
style.use('ggplot')
df = pd.read_csv('TSLA.csv', parse_dates=True, index_col=0)
df_ohlc = df['Close'].resample('10D').ohlc()
df_volume = df['Volume'].resample('10D').sum()
df_ohlc = df_ohlc.reset_index()
df_ohlc['Date'] = df_ohlc['Date'].map(mdates.date2num)
print (df_ohlc.tail())
ax1 = plt.subplot2grid((6,1), (0,0), rowspan=5, colspan=1)
ax2 = plt.subplot2grid((6,1), (5,0), rowspan=1, colspan=1,sharex=ax1)
candlestick_ohlc(ax1, df_ohlc.values, width=2, colorup='g')
ax2.fill_between(df_volume.index.map(mdates.date2num),df_volume.values,0)
plt.show()
def plot(startday, islog = False):
columns = ['date','open','high','low','close','unknown','volume']
filename_full = 'data/'+symbol+'_full_'+contractType+'_future.csv'
filename_matrix = 'data/'+symbol+'_matrix_full_'+contractType+'_future.csv'
filename_pred = 'data/'+symbol+'_pred_'+contractType+'_future.csv'
filename_back = 'data/'+symbol+'_back_'+contractType+'_future.csv'
filename_fulltemp = 'data/'+symbol+'_fulltemp_'+contractType+'_future.csv'
filename_matrix_temp = 'data/'+symbol+'_matrix_fulltemp_'+contractType+'_future.csv'
now = datetime.now()
now = time.mktime(now.timetuple())
if os.path.exists(filename_full):
with open(filename_full, 'r') as f:
df = pd.read_csv(f,header=0, index_col=0, sep ='\t')
with open(filename_pred,'r') as f_pred:
df_pred = pd.read_csv(f_pred, header=0, index_col=0, sep='\t')
with open(filename_matrix,'r') as f_matrix:
df_matrix = pd.read_csv(f_matrix, header = 0,index_col=0,sep='\t')
with open(filename_back,'r') as f_back:
df_back = pd.read_csv(f_back, header = 0,index_col=0,sep='\t')
with open(filename_fulltemp,'r') as f_temp:
df_concat = pd.read_csv(f_temp, header = 0,index_col=0,sep='\t')
with open(filename_matrix_temp,'r') as f_matrix_temp:
df_matrix_temp = pd.read_csv(f_matrix_temp, header = 0,index_col=0,sep='\t')
df = df.drop(['unknown'],axis = 1)
df_pred = df_pred.fillna(0)
df_back = df_back.fillna(0)
# df_matrix_temp = df_matrix_temp.fillna(0)
df_matrix_temp['flag'] = df_matrix_temp['flag'].apply(lambda x:x*1.1)
df_matrix_temp['date1'] = df_matrix_temp['date']
df_back['shortposition'] = df_back['shortposition'].apply(lambda x:-x)
df['date'] = df['date'].apply(lambda x:mdates.date2num(datetime.fromtimestamp(x)))
df_matrix['date'] = df_matrix['date'].apply(lambda x:mdates.date2num(datetime.fromtimestamp(x)))
df_pred['date'] = df_pred['date'].apply(lambda x:mdates.date2num(datetime.fromtimestamp(x)))
df_back['date'] = df_back['date'].apply(lambda x:mdates.date2num(datetime.fromtimestamp(x)))
df_concat['date'] = df_concat['date'].apply(lambda x:mdates.date2num(datetime.fromtimestamp(x)))
df_matrix_temp['date'] = df_matrix_temp['date'].apply(lambda x:mdates.date2num(datetime.fromtimestamp(x)))
#df['date'] = pd.to_datetime(df['date'])
#df['date'] = pd.to_datetime(df['date'])
#df['date'] = df['date'].values.astype('float64')
#df['date'].astype('float64')#['date'].apply(lambda x:np.ndarray.astype(x/1000))
ohlc = [tuple(x) for x in df_concat.to_records(index=False)]
ohlc = ohlc[::5]
#fig = plt.figure()
ax1 = plt.subplot2grid((9,1), (0,0), rowspan = 5)
plt.ylabel('Price')
plt.title('BTC_future_next_quarter')
candlestick_ohlc(ax1, ohlc, width=0.0004, colorup='#77d879', colordown='#db3f3f')
xrange1 = mdates.date2num(datetime.fromtimestamp(now-startday*24*3600))
xrange2 = mdates.date2num(datetime.fromtimestamp(now+0.5*24*3600))
ax1.scatter('date', 'longposition',color='purple',s = 0.18, data=df_back,label = 'long-pos')
ax1.scatter('date', 'shortposition',color='black',s = 0.18, data=df_back,label = 'short-pos')
ax1.set_xlim(xrange1, xrange2)
price_low = df_matrix_temp['close'][df_matrix_temp['date1'].searchsorted(now-startday*24*3600)[0]:].min()*(1-0.08)
price_high = df_matrix_temp['close'][df_matrix_temp['date1'].searchsorted(now-startday*24*3600)[0]:].max()*(1+0.08)
ax1.set_ylim(price_low, price_high)
ax1.xaxis.set_major_formatter(mdates.DateFormatter('%m-%d %H:%M'))
ax1.xaxis.set_major_locator(mticker.MaxNLocator(15))
ax1.xaxis.set_minor_locator(mticker.MaxNLocator(5))
ax1.legend(loc='best',prop={'size':10})
ax1.grid(True)
ax2 = plt.subplot2grid((9,1), (5,0), rowspan = 1, sharex = ax1)
ax2.scatter('date', 'flag', color='red',s=0.1,data=df_matrix_temp,label='history')
for model in models:
ax2.plot('date', 'prob_'+model, linewidth = 0.4,data=df_pred,label='prediction_'+model)
ax2.set_xlim(xrange1, xrange2)
ax2.set_ylim(-1.25,1.25)
ax2.legend(loc='best',prop={'size':6})
ax2.grid(True)
ax3 = plt.subplot2grid((9,1), (6,0), rowspan = 3, sharex = ax1)
ax3.plot('date','btc_spot_back',data=df_back,label='spot_back')
ax3.plot('date','btc_spot_hist',data=df_back,label='spot_history')
ax3.plot('date','btc_future_back_RF',data=df_back,label='future_back_RF')
ax3.plot('date','btc_future_back_XGB',data=df_back,label='future_back_XGB')
ax3.plot('date','btc_future_hist',data=df_back,label='future_history')
ax3.set_xlim(xrange1, xrange2)
ax3.legend(loc='best',prop={'size':8})
if islog:
ax3.set_yscale('log')
ax3.grid(True)
for label in ax3.xaxis.get_ticklabels():
label.set_rotation(45)
plt.xlabel('Date')
plt.setp(ax1.get_xticklabels(), visible=False)
plt.setp(ax2.get_xticklabels(), visible=False)
#plt.setp(ax3.get_xticklabels(), visible=False)
#plt.legend()
plt.subplots_adjust(left=0.09, bottom=0.2, right=0.94, top=0.90, wspace=0.12, hspace=0)
#plt.show()
plt.savefig('data/figure.png')
