In [152]:
import pandas as pd
import seaborn as sns
from IPython.core.interactiveshell import InteractiveShell
InteractiveShell.ast_node_interactivity = "all"
In [153]:
df = pd.read_csv("/home/vb/code/gl/viblo.se/posts/.devdrive/log-20250913.csv")
df.head()
Out[153]:
| env | git clone | go build | pollycore restore | pollycore build | |
|---|---|---|---|---|---|
| 0 | dev | 9.396707 | 9.027582 | 1.252155 | 3.132748 |
| 1 | dev | 8.977330 | 8.949979 | 0.952710 | 1.847611 |
| 2 | dev | 8.182659 | 8.908458 | 0.925588 | 1.836672 |
| 3 | dev | 8.189944 | 9.835053 | 0.984734 | 1.854889 |
| 4 | dev | 8.581076 | 9.358740 | 0.980548 | 1.819684 |
In [154]:
df.describe()
Out[154]:
| git clone | go build | pollycore restore | pollycore build | |
|---|---|---|---|---|
| count | 80.000000 | 80.000000 | 80.000000 | 80.000000 |
| mean | 8.666601 | 8.886130 | 1.073471 | 2.043051 |
| std | 0.684723 | 1.051348 | 0.378670 | 0.700125 |
| min | 7.681987 | 6.819397 | 0.897706 | 1.773370 |
| 25% | 8.188122 | 8.299738 | 0.934294 | 1.825259 |
| 50% | 8.547404 | 9.252857 | 0.970038 | 1.846967 |
| 75% | 8.912638 | 9.516928 | 1.019691 | 1.914963 |
| max | 10.733647 | 11.472750 | 3.595697 | 7.479691 |
In [155]:
def remove_worst_outliers(df, column):
Q1 = df[column].quantile(0.25)
Q3 = df[column].quantile(0.75)
IQR = Q3 - Q1
upper_bound = Q3 + 2 * IQR
return df[df[column] <= upper_bound]
for col in ['git clone', 'go build', 'pollycore restore', 'pollycore build']:
df = remove_worst_outliers(df, col)
In [156]:
df.groupby('env').mean()
df.groupby('env').median()
df.groupby('env').min()
Out[156]:
| git clone | go build | pollycore restore | pollycore build | |
|---|---|---|---|---|
| env | ||||
| dev | 8.355994 | 9.396100 | 0.946092 | 1.838001 |
| noav | 8.892655 | 8.950315 | 0.943635 | 1.840549 |
| ntfs | 8.740263 | 9.744776 | 0.960335 | 1.925375 |
| wsl | 7.989642 | 7.245553 | 1.010941 | 1.840639 |
Out[156]:
| git clone | go build | pollycore restore | pollycore build | |
|---|---|---|---|---|
| env | ||||
| dev | 8.252210 | 9.356297 | 0.935502 | 1.842094 |
| noav | 8.857984 | 8.930260 | 0.939725 | 1.830228 |
| ntfs | 8.556161 | 9.707799 | 0.956464 | 1.911097 |
| wsl | 7.996752 | 7.251345 | 0.998978 | 1.841612 |
Out[156]:
| git clone | go build | pollycore restore | pollycore build | |
|---|---|---|---|---|
| env | ||||
| dev | 8.164013 | 8.908458 | 0.918885 | 1.795018 |
| noav | 8.544622 | 8.226923 | 0.902623 | 1.791978 |
| ntfs | 8.449004 | 9.419567 | 0.897706 | 1.861951 |
| wsl | 7.681987 | 6.819397 | 0.963385 | 1.782720 |
In [157]:
dfm = df.groupby('env').mean()
(-100*(dfm.sub(dfm.loc['ntfs'], axis=1) / dfm.loc['ntfs'])).round()
dfm = df.groupby('env').median()
(-100*(dfm.sub(dfm.loc['ntfs'], axis=1) / dfm.loc['ntfs'])).round()
Out[157]:
| git clone | go build | pollycore restore | pollycore build | |
|---|---|---|---|---|
| env | ||||
| dev | 4.0 | 4.0 | 1.0 | 5.0 |
| noav | -2.0 | 8.0 | 2.0 | 4.0 |
| ntfs | -0.0 | -0.0 | -0.0 | -0.0 |
| wsl | 9.0 | 26.0 | -5.0 | 4.0 |
Out[157]:
| git clone | go build | pollycore restore | pollycore build | |
|---|---|---|---|---|
| env | ||||
| dev | 4.0 | 4.0 | 2.0 | 4.0 |
| noav | -4.0 | 8.0 | 2.0 | 4.0 |
| ntfs | -0.0 | -0.0 | -0.0 | -0.0 |
| wsl | 7.0 | 25.0 | -4.0 | 4.0 |
In [158]:
df2 = df.melt(id_vars=['env'], value_vars=['git clone','go build','pollycore restore','pollycore build'], var_name='action', value_name='time')
In [159]:
sns.set_context("notebook")
import matplotlib.pyplot as plt
for action in df2.action.unique():
sns.catplot(data=df2[(df2.action == action)], hue="env", x="time", col='action', kind='box', aspect=1.5)
plt.savefig(action + ".png")
Out[159]:
<seaborn.axisgrid.FacetGrid at 0x718c50f55250>
Out[159]:
<seaborn.axisgrid.FacetGrid at 0x718c50db5be0>
Out[159]:
<seaborn.axisgrid.FacetGrid at 0x718c50bdaa80>
Out[159]:
<seaborn.axisgrid.FacetGrid at 0x718c5389f140>
