Heatmap only has annotation text in the top row only

[pybokeh]

Example to duplicate the problem:

# Required libraries 
import seaborn as sns
import matplotlib.pyplot as plt

# Load the iris dataset into a Pandas dataframe
iris_data = sns.load_dataset('iris')

# Creating the correlation matrix of the iris dataset
iris_corr_matrix = iris_data.corr(numeric_only=True)
print(iris_corr_matrix)

# Create the heatmap using the `heatmap` function of Seaborn
sns.heatmap(iris_corr_matrix, cmap='coolwarm', annot=True)

# Display the heatmap using the `show` method of the `pyplot` module from matplotlib.
plt.show()

My environment:
Python 3.11.4 virtual environment via venv module, everything installed with pip
OS: Ubuntu 22.04
Seaborn version: 0.12.2
Pandas version: 2.1.0
MATPLOTLIB version: 3.8.0

My pip list output:
Package Version

---

asttokens 2.4.0
atpublic 4.0
backcall 0.2.0
bidict 0.22.1
certifi 2023.7.22
charset-normalizer 3.2.0
comm 0.1.4
contourpy 1.1.1
cycler 0.11.0
debugpy 1.8.0
decorator 5.1.1
duckdb 0.8.1
duckdb-engine 0.9.2
executing 1.2.0
filelock 3.12.4
fonttools 4.42.1
greenlet 2.0.2
ibis-framework 6.2.0
idna 3.4
ipykernel 6.25.2
ipython 8.15.0
jedi 0.19.0
jupyter_client 8.3.1
jupyter_core 5.3.1
kiwisolver 1.4.5
markdown-it-py 3.0.0
matplotlib 3.8.0
matplotlib-inline 0.1.6
mdurl 0.1.2
multipledispatch 1.0.0
nest-asyncio 1.5.7
numpy 1.26.0
packaging 23.1
pandas 2.1.0
parso 0.8.3
parsy 2.1
patsy 0.5.3
pexpect 4.8.0
pickleshare 0.7.5
Pillow 10.0.1
pip 23.2.1
platformdirs 3.10.0
pooch 1.7.0
prompt-toolkit 3.0.39
psutil 5.9.5
ptyprocess 0.7.0
pure-eval 0.2.2
pyarrow 12.0.1
Pygments 2.16.1
pyparsing 3.1.1
python-dateutil 2.8.2
pytz 2023.3.post1
pyzmq 25.1.1
requests 2.31.0
rich 13.5.2
scipy 1.11.2
seaborn 0.12.2
setuptools 65.5.0
six 1.16.0
SQLAlchemy 2.0.20
sqlalchemy-views 0.3.2
sqlglot 17.16.2
stack-data 0.6.2
statsmodels 0.14.0
toolz 0.12.0
tornado 6.3.3
tqdm 4.66.1
traitlets 5.10.0
typing_extensions 4.7.1
tzdata 2023.3
urllib3 2.0.4
wcwidth 0.2.6
xxhash 3.3.0

[pybokeh]

Looks like problem goes away if I downgrade MATPLOTLIB to version 3.7.3.

[mwaskom]

Hm, I cannot replicate

So maybe this has been already addressed on master, although nothing immediately comes to mind.

[stefanistrate]

I am having the same issue with seaborn 0.12.2 from PyPI, but things seem to be work just fine with the github master branch. Maybe it's worth pushing a new version to PyPI soon?

[mwaskom]

It would be helpful for someone to track down the root cause here and determine if it needs to be reported upstream to matplotlib as a regression, independent of whether seaborn already has a workaround for it on the dev branch.

[thuiop]

I'll try to take a look.

[thuiop]

So the issue is that the value returned by pcolormesh is now a different class, as mentioned here. The get_array method no longer returns a flattened array by default.
This has been fixed by @ksunden in this commit b3e6b8d.

[ylevental]

Looks like problem goes away if I downgrade MATPLOTLIB to version 3.7.3.

Downgrading worked for me too. It seems that this problem will be fixed soon in the pip release.

[mwaskom]

Closing this as the issue is resolved in seaborn v0.13.0 which will be released in the next day or two barring any last-minute surprises about the RC.

[mwaskom]

I am guessing you updated seaborn in a different environment, or perhaps forgot to restart your kernel:

[engnadeau]

I am guessing you updated seaborn in a different environment, or perhaps forgot to restart your kernel:

yeah, caught my error but you were too fast 🤣

[rv-12]

Looks like problem goes away if I downgrade MATPLOTLIB to version 3.7.3.

worked for me

[paulinavita]

Hi, I wonder why it's not working for me, I downgraded the matplotlib and update seaborn to latest already. Also restarted my kernel

I still only got the first row of the heatmap showing

My environment:
Python 3.11.4, everything installed with pip
OS: Mac OS Ventura 13.2.1
Seaborn version: 0.13.1
Pandas version: 2.2.1
MATPLOTLIB version: 3.7.3

Any step I am missing? Thank you

[mwaskom]

I would confirm that you actually have those versions imported in the python process where you’re making the plot

[paulinavita]

Ah, sorry about that. Yes, I confirm it works now. Thank you