Plotting rectangle with the color column Matplotlib Python

Question

I was trying to draw the color of the candles of OHLC. That was accomplished using the code available at: https://stackoverflow.com/a/51417197/4948889

But there is issue while I am trying it in my code. I will let know what situation is here.

dataset_train = pd.read_csv('EURUSD_M1TechnicalIndicators.csv',usecols=[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15])
dataset_train.head(10)
dataset_test = pd.read_csv('Test_EURUSD_M1TechnicalIndicators.csv',usecols=[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15],nrows=50)
dataset_test.head(10)

train  = dataset_train.filter(['open','high','low','close','closeTarget',"k","d","atr","macdmain","macdsgnal","bbup","bbmid","bblow"], axis=1)
x = train.values #returns a numpy array
min_max_scaler = preprocessing.MinMaxScaler()
x_scaled = min_max_scaler.fit_transform(x)
datasetTrain = pd.DataFrame(x_scaled)
datasetTrain.columns = ['open','high','low','close','closeTarget',"k","d","atr","macdmain","macdsgnal","bbup","bbmid","bblow"]
datasetTrain.head(10) 

test = dataset_test.filter(['open','high','low','close','closeTarget',"k","d","atr","macdmain","macdsgnal","bbup","bbmid","bblow"], axis=1)
x = test.values #returns a numpy array
min_max_scaler = preprocessing.MinMaxScaler()
x_scaled = min_max_scaler.fit_transform(x)
datasetTest = pd.DataFrame(x_scaled)
datasetTest.columns = ['open','high','low','close','closeTarget',"k","d","atr","macdmain","macdsgnal","bbup","bbmid","bblow"]
datasetTest.head(10)

plt.figure(figsize=(25,5))
plt.plot(xTrain[:,3])
plt.title('Train (' +str(len(xTrain))+' data points)')
plt.show()
plt.figure(figsize=(10,3))
plt.plot(xTest[:,0])
plt.title('Test (' +str(len(xTest))+' data points)')
plt.show()

The output till here was:

Then I tried this:

def draw_rects(ax, quotes, width=5., height=1., yloc=1., colorup='g', 
               colordown='r', edgecolor='k', alpha=1.0):

    OFFSET = width / 2.0
    patches = []
    for q in quotes:
        t, open, close, high, low = q[:5]
        if close > open:
            color = colorup
        else:
            color = colordown

        rect = Rectangle(
            xy=(t - OFFSET, yloc),
            width=width,
            height=height,
            facecolor=color,
            edgecolor=edgecolor,
        )
        rect.set_alpha(alpha)
        patches.append(rect)
        ax.add_patch(rect)

    ax.autoscale_view()

    return patches

fig, ax = plt.subplots(1,1)
quotes = xTest
p1 = draw_rects(ax, xTrain, yloc=1)
p2 = draw_rects(ax, xTest, yloc=4)
labels = [item.get_text() for item in ax.get_yticklabels()]
labels[2] = 'Train'
labels[8] = 'Test'
ax.set_yticklabels(labels)
plt.show()

Output:

I was expecting the candles color to be shown. So that above process didn't worked for me. So I created another column in the dataset as color and shiftedcolor

Now I am thinking to just showcase the color data from the datasets in the candle fashion using the rectangles. Please help me in doing that.

Here are the datasets used by me. EURUSD_M1TechnicalIndicators And Test_EURUSD_M1TechnicalIndicators

By looking at the plot now knowing that your Files have 30202 lines - couldn't this plot be ok but just hard to read because of thousands of (possibly overlapping) rectangles? — SpghttCd
– SpghttCd, Commented Jul 19, 2018 at 21:38

SpghttCd · Accepted Answer · 2018-07-20 06:23:04Z

1

I don't really get the complexity of your problem plotting some rectangles (no offending, please correct me if I just ignored important things, in other words: tl;dr...)

But just as an example, how I would plot those two rows (let's say as a different approach to start discussing if it's worth it...):

x = np.arange(10)
y = np.ones(10)

bools = np.random.randint(0, 2, 10).astype(bool)
colors = np.array(bools, str)
colors[bools] = 'g'
colors[~bools] = 'r'

rx = 1
ry = 2
rect = [(-rx, -ry), (rx, -ry), (rx, ry), (-rx, ry)]

plt.figure()
plt.scatter(x, y, facecolor=colors, verts=rect, s=1000)
plt.scatter(x, y+3, facecolor=colors[::-1], verts=rect, s=1000)
plt.ylim(0, 5)
plt.yticks([1, 4], ['Train', 'Test'])

leads to:

EDIT: The same thing applied to your data files: (And I saw you want also black for equal values, so added this, too)

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

dataset_train = pd.read_csv('EURUSD_M1TechnicalIndicators.txt', usecols=['open', 'close'])
dataset_test = pd.read_csv('Test_EURUSD_M1TechnicalIndicators.csv', usecols=['open', 'close'])

gt_train = (dataset_train.open - dataset_train.close) > 0
gt_test = (dataset_test.open - dataset_test.close) > 0
eq_train = (dataset_train.open - dataset_train.close) == 0
eq_test = (dataset_test.open - dataset_test.close) == 0

y_train = np.ones(len(gt_train))
y_test = np.ones(len(gt_test))

colors_train = np.array(gt_train, str)
colors_test = np.array(gt_test, str)
colors_train[gt_train] = 'g'
colors_train[~gt_train] = 'r'
colors_train[eq_train] = 'k'
colors_test[gt_test] = 'g'
colors_test[~gt_test] = 'r'
colors_test[eq_test] = 'k'

rx = .2
ry = 2
rect = [(-rx, -ry), (rx, -ry), (rx, ry), (-rx, ry)]

plt.figure()
plt.scatter(np.arange(y_train.size), y_train, facecolor=colors_train, verts=rect, s=1000)
plt.scatter(np.arange(y_test.size), y_test+3, facecolor=colors_test, verts=rect, s=1000)
plt.ylim(0, 5)
plt.yticks([1, 4], ['Train', 'Test'])

edited Jul 20, 2018 at 6:23

answered Jul 19, 2018 at 13:14

SpghttCd

10.9k2 gold badges23 silver badges28 bronze badges

Sign up to request clarification or add additional context in comments.

7 Comments

Jaffer Wilson Over a year ago

Wow, this one is what an excellent answer I have got till now. Can you just help me get them labelled please?

SpghttCd Over a year ago

Which labels please? You can set 'x = np.arange(-2, 4)' if you like to get the x tick labels you posted above. Or 'plt.yticks([1, 4], ['Train', 'Test'])' for your y axis labels. Or did you want to label the markers themselves?

Jaffer Wilson Over a year ago

It was obviously the axis label what I mean. I guess is done. One more thing, Can you tell me if I wanted to increase the gap between the two rectangles then what I have to do?

SpghttCd Over a year ago

You can change the aspect ratio of the rectangle which is defined in the verts kwarg. I added two parameters rxand ry, so that you can easily adjust it.

Jaffer Wilson Over a year ago

Thank you sir for your help. I am checking it out.

|

Collectives™ on Stack Overflow

Plotting rectangle with the color column Matplotlib Python

1 Answer 1

7 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

7 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related