Don't have the original email, hence replying this way. On Sat, Feb 20, 2021 at 12:13:30PM +0100, jak wrote: > Il 20/02/2021 01:56, Dino ha scritto: > > > > trying to do some dayaviz with Italian Covid Open Data ( > > https://github.com/italia/covid19-opendata-vaccini/ ) > > > > here's how I pull my data: > > ____________________________ > > import sys > > import urllib.request > > import pandas as pd > > import ssl > > ssl._create_default_https_context = ssl._create_unverified_context > > > > URL = > > "https://github.com/italia/covid19-opendata-vaccini/blob/master/dati/somministrazioni-vaccini-latest.csv?raw=true"; > > > > > > with urllib.request.urlopen(URL) as url: > > df = pd.read_csv(url) > > ____________________________ > > > > One of my diagrams came out screwed up today, and I am having a hard > > time understanding what went wrong: > > > > https://imgur.com/a/XTd4akn > > > > Any ideas? > > > > Thanks > > > I don't think this is the cause of your problem and in addition I don't > know about pandas. In any case you send pandas some records that contain > the date in string format and this gives the alphabetic continuity but > if the data contained time holes, these would not be represented in your > graph. Maybe you should add an intermediate step and convert strings > dates to datetime format before you create the chart (but perhaps pandas > takes care of this. I don't know this).
Pretty much what he said... Parse the dates. Oh and you generally don't need the dance with urllib, pandas can do that for you. ``` data_url = r"https://github.com/italia/covid19-opendata-vaccini/blob/master/dati/somministrazioni-vaccini-latest.csv?raw=true"; df = pd.read_csv(data_url, parse_dates=True, index_col="data_somministrazione") plt.figure(figsize=(15,10)) plt.xticks(rotation=70) sns.lineplot(x=df.index, y="prima_dose", data=df, hue="nome_area", ci=None) ``` Yields: https://labrat.space/irc/3b0be1f11e6c687b/download%20(1).png Cheers, Reto -- https://mail.python.org/mailman/listinfo/python-list
