Is there an existing issue for this?

• I have searched the existing issues

Bug description

Hi, I am trying the demo . I used conda to install DeepLabCut.

However, I got this error of not finding cmap in matplotlib when I finished training and ran the evaluation deeplabcut.evaluate_network(path_config_file,plotting=True):

Results for 10000  training iterations: 95 1 train error: 4.09 pixels. Test error: 5.71  pixels.
With pcutoff of 0.4  train error: 4.09 pixels. Test error: 5.71 pixels
Thereby, the errors are given by the average distances between the labels by DLC and the scorer.
Plotting...
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/chz31/DeepLabCut/deeplabcut/pose_estimation_tensorflow/core/evaluate.py", line 1019, in evaluate_network
    Plotting(
  File "/home/chz31/DeepLabCut/deeplabcut/pose_estimation_tensorflow/core/evaluate.py", line 201, in Plotting
    colors = visualization.get_cmap(len(comparisonbodyparts), name=cfg["colormap"])
  File "/home/chz31/DeepLabCut/deeplabcut/utils/visualization.py", line 36, in get_cmap
    return plt.cm.get_cmap(name, n)
AttributeError: module 'matplotlib.cm' has no attribute 'get_cmap'

After activating conda activate DEEPLABCUT, the Python used is 3.9. 19 in the conda environment. In the base environment, matplotlib.cm.cmap still exists with Python 3.11 but the method appeared to be deprecated.

Tensorflow 2.10 also could not find CUDA. The CUDA installed in my machine is 11.8, which appears to not be compatible with Tensorflow, but I think it's not relevant to this issue.

Operating System

WSL2 (Ubuntu) under Windows 11

DeepLabCut version

dlc version 2.3.9

DeepLabCut mode

single animal

Device type

gpu NVIDIA Geforce GTX 1070 Max Q
nvidia-smi returns NVIDIA-SMI 550.76.01 Driver Version: 552.22 CUDA Version: 11.8

Steps To Reproduce

No response

Relevant log output

No response

Anything else?

No response

Code of Conduct

• I agree to follow this project's Code of Conduct