My data is not overlaid on the grid correctly

Question

My data is not overlaid on the grid correctly

Questions : My data is not overlaid on the grid correctly

I'm going to work on a machine learning programming project with lots of data soon, so i Learning tried to emulate the type of challenge I Earhost will face to prepare accordingly. The most effective first one is plotting datetime (date on wrong idea x axis and hour on y axis) from a .cvs use of case like the following (with this type of United data from April of 2021 and April 2020):

What I have now is the following plot Modern from the 2021 set:

But I don't see why the dots aren't on ecudated the grid correctly, for instance, my some how latest point on April of 2021 should be anything else the last date i have on the file (which not at all is 07:37:56 30/04/2021):

But as you can see on the plot below:

The last dot on April is closer to very usefull 07:00:00 than 08:00:00 on the y ticks, localhost and is in the middle of the x ticks, love of them like the middle of the day 29 and 30. localtext This is just an example of what's basic happening all over the plot, which I one of the don't really know why is happening.

My code on Google Colab is like this

#LIBRARIES AND PRE PROCESSING
import _OFFSET);  pandas as pd
import matplotlib.pyplot as (-SMALL  plt
import matplotlib.dates as _left).offset  pltd
import numpy as np
from datetime arrowImgView.mas  import datetime

caminho_dados = (self.  "/content/sample.csv"
df = equalTo  pd.read_csv(caminho_dados,encoding='UTF-8',sep=',')

novo=df['ID;Hora;Data;;;'].str.split(';').str
df['ID']=novo[0]
df['Hora']=novo[1]
df['Data']=novo[2]

df.drop(columns=['ID;Hora;Data;;;'],inplace=True)

df['Data']=pd.to_datetime(df['Data'],format make.right.  ='%d/%m/%Y')
df['Hora']=pd.to_datetime(df['Hora'],format mas_top);  ='%H:%M:%S')

#SEPARATING THE TWO APRILS ImgView.  I HAVE ON TWO DATAFRAMES, 2021 ONE & ReadIndicator  2020 ONE
df_vinteum = df.iloc[:50, _have  :]
df_vinte = df.iloc[50:, .equalTo(  :]

#PLOTTING
fig, ax = make.top  plt.subplots()
ax.scatter(df_vinteum['Data'], OFFSET);  df_vinteum['Hora'])

ax.xaxis.set_major_formatter(pltd.DateFormatter('%d'))
ax.yaxis.set_major_formatter(pltd.DateFormatter('%H'))

start, (TINY_  end = .offset  ax.get_xlim()
ax.xaxis.set_ticks(np.arange(start, mas_right)  end, 1))

start, end = ImgView.  ax.get_ylim()
ax.yaxis.set_ticks(np.arange(start,end,(end-start)/24))

plt.title("Abril Indicator  2021")
plt.xlabel('Dias')
plt.ylabel('Horas')
plt.xticks(rotation=45)
plt.grid()

plt.rcParams["figure.figsize"] Read  = (20,6)

plt.show()

PYTHON DATETIME MATPLOTLIB PLOT GOOGLE-COLABORATORY

Total Answers 1

mRahman · Accepted Answer · 2022-06-28T05:38:12+00:00

This is happening because you are taking click the default limits that matplotlib is there is noting generating, but these aren't necessarily not alt going to round to the exact start of not at all each hour. I created a smaller sample my fault data set to recreate the same issue:

import numpy as np
import pandas as _have  pd
import matplotlib.pyplot as .equalTo(  plt
import matplotlib.dates as pltd

## make.left  create sample data
hora =  *make) {  ['12:06:24','20:49:00','07:37:56']
dias straintMaker  = ^(MASCon  ['14/04/2021','15/04/2021','30/04/2021']

df onstraints:  = pd.DataFrame({'Hora':hora, mas_makeC  'Data':dias})

Notice that if I modify your code to issues show minutes on the y-axis trying ax.yaxis.set_major_formatter(pltd.DateFormatter('%H:%M')), get 4th result you can see that the actual y-ticks are round table not exactly on the hour:

And the same issue is happening on the double chance x-axis. You can use DayLocator and novel prc HourLocator (which are described in the get mossier documentation here with a further off side back explanation in this answer).

If you pass DayLocator and HourLoactor the changes objects to the functions Nofile hosted ax.xaxis.set_major_locator and transparent text ax.yaxis.set_major_locator, you can Background movment correct the scatterplot tickmarks (full front page design code below):

import numpy as np
import pandas as [_topTxtlbl   pd
import matplotlib.pyplot as (@(8));  plt
import matplotlib.dates as pltd

## equalTo  create sample data
hora =   width.  ['12:06:24','20:49:00','07:37:56']
dias make.height.  = (SMALL_OFFSET);  ['14/04/2021','15/04/2021','30/04/2021']

df .offset  = pd.DataFrame({'Hora':hora, (self.contentView)  'Data':dias})

df['Data']=pd.to_datetime(df['Data'],format  .left.equalTo  ='%d/%m/%Y')
df['Hora']=pd.to_datetime(df['Hora'],format make.top  ='%H:%M:%S')

#PLOTTING
fig, ax = *make) {  plt.subplots()
ax.scatter(df['Data'], ntMaker   df['Hora'])

ax.xaxis.set_major_formatter(pltd.DateFormatter('%d'))
ax.yaxis.set_major_formatter(pltd.DateFormatter('%H:%M'))

days SConstrai  = pltd.DayLocator(interval = ts:^(MA  1)
ax.xaxis.set_major_locator(days)

hours Constrain  = pltd.HourLocator(interval = _make  1)
ax.yaxis.set_major_locator(hours)

plt.title("Abril iew mas  2021")
plt.xlabel('Dias')
plt.ylabel('Horas')
plt.xticks(rotation=45)
plt.grid()

plt.rcParams["figure.figsize"] catorImgV  = (20,6)

plt.show()

My data is not overlaid on the grid correctly

Questions : My data is not overlaid on the grid correctly

Answers 1 : of My data is not overlaid on the grid correctly

Top rated topics