在散点图中注释 csv 列

Question

I have two dataset in csv format:

df2

type  prediction          100000     155000    
 0           0            2.60994   3.40305
 1           1           10.82100  34.68900
 0           0            4.29470   3.74023
 0           0            7.81339   9.92839
 0           0           28.37480  33.58000

df

 TIMESTEP   id  type         y         z         v_acc
  100000   8054     1     -0.317192 -0.315662   15.54430
  100000    669     0      0.352031 -0.008087   2.60994 
  100000    520     0      0.437786  0.000325   5.28670
  100000   2303     1      0.263105  0.132615   7.81339 
  105000   8055     1      0.113863  0.036407   5.94311

I am trying to match value of df2[100000] to df1[v_acc] . If value matched, I am making scatter plot from df with columns y and z . After that I want to to annoted scatter point with matched value.

What I want is:

(I want all annotaions in a same plot).

I tried to code in python for such condition but I am not getting all annotation points in a single plot instead I am getting multi plots with a single annotation. I am also getting this error:

TypeError                                 Traceback (most recent call last)
File /Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/IPython/core/formatters.py:339, in BaseFormatter.__call__(self, obj)
    337     pass
    338 else:
--> 339     return printer(obj)
    340 # Finally look for special method names
    341 method = get_real_method(obj, self.print_method)

File /Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/IPython/core/pylabtools.py:151, in print_figure(fig, fmt, bbox_inches, base64, **kwargs)
    148     from matplotlib.backend_bases import FigureCanvasBase
    149     FigureCanvasBase(fig)
--> 151 fig.canvas.print_figure(bytes_io, **kw)
    152 data = bytes_io.getvalue()
    153 if fmt == 'svg':

File /Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/matplotlib/backend_bases.py:2295, in FigureCanvasBase.print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, bbox_inches, pad_inches, bbox_extra_artists, backend, **kwargs)
   2289     renderer = _get_renderer(
   2290         self.figure,
   2291         functools.partial(
   2292             print_method, orientation=orientation)
   2293     )
   2294     with getattr(renderer, "_draw_disabled", nullcontext)():
-> 2295         self.figure.draw(renderer)
   2297 if bbox_inches:
...
    189 if len(self) == 1:
    190     return converter(self.iloc[0])
--> 191 raise TypeError(f"cannot convert the series to {converter}")

TypeError: cannot convert the series to <class 'float'>

Can I get some help to make a plot as I want?

Thank you. My code is here:

df2 = pd.read_csv('./result.csv')
print(df2.columns)
#print(df2.head(10))
df  = pd.read_csv('./main.csv')
df = df[df['TIMESTEP'] == 100000]

for i in df['v_acc']:
    for j in df2['100000']:
        # sometimes numbers are long and different after decimals.So mathing 0.2f only
        if "{0:0.2f}".format(i) == "{0:0.2f}".format(j):
            plt.figure(figsize = (10,8))
            sns.scatterplot(data = df, x = "y", y = "z", hue = "type", palette=['red','dodgerblue'], legend='full')
            plt.annotate(i, (df['y'][df['v_acc'] == i], df['z'][df['v_acc'] == i]))
            plt.grid(False)
            plt.show()
            break

Answer 1

the reason for the multiple plots is because are you using plt.figure() inside the loop. This will create a single figure for each loop. You need to create that outside and only the individual scatter and annotate within the loop. Here is the updated code that ran for the data you provided. Other than that, think your code is fine...

fig, ax=plt.subplots(figsize = (7,7)) ### Keep this before the loop and call it as subplot
for i in df['v_acc']:
    for j in df2[100000]:
        # sometimes numbers are long and different after decimals.So mathing 0.2f only
        if "{0:0.2f}".format(i) == "{0:0.2f}".format(j):
            #plt.figure(figsize = (10,8))
            ax=sns.scatterplot(data = df, x = "y", y = "z", hue = "type", palette=['red','dodgerblue'], legend='full')
            ax.annotate(i, (df['y'][df['v_acc'] == i], df['z'][df['v_acc'] == i]))
            break

plt.grid(False)  ### Keep these two after the loop, just one show for one plot
plt.show()

Output plot