Error with installation

[joanetLiew]

Hi @justin-delano @RachelSilverstein ,

Thanks for developing this valuable software for the community.
However, after installing it following the instruction, i ran into an error, which seems to be sklearn relevant.
I could not figure out how to fix it after trying multiple different versions of sklearn.
Can you please help me out of this?

python make_predictions.py -h
2025-04-24 14:28:26.651427: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: SSE4.1 SSE4.2 AVX AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
WARNING:tensorflow:SavedModel saved prior to TF 2.5 detected when loading Keras model. Please ensure that you are saving the model with model.save() or tf.keras.models.save_model(), *NOT* tf.saved_model.save(). To confirm, there should be a file named "keras_metadata.pb" in the SavedModel directory.
2025-04-24 14:28:28.537732: I tensorflow/core/common_runtime/process_util.cc:146] Creating new thread pool with default inter op setting: 2. Tune using inter_op_parallelism_threads for best performance.
/home/designer/.local/lib/python3.8/site-packages/sklearn/base.py:348: InconsistentVersionWarning: Trying to unpickle estimator OneHotEncoder from version 0.23.1 when using version 1.3.2. This might lead to breaking code or invalid results. Use at your own risk. For more info please refer to:
https://scikit-learn.org/stable/model_persistence.html#security-maintainability-limitations
  warnings.warn(
/home/designer/.local/lib/python3.8/site-packages/sklearn/base.py:458: UserWarning: X has feature names, but OneHotEncoder was fitted without feature names
  warnings.warn(
Traceback (most recent call last):
  File "make_predictions.py", line 54, in <module>
    muts_df = predict_from_saved_model(saved_model_dir, muts_df, DATA_MEAN, DATA_STD, pams)
  File "/home/designer/sw/structuralBioinformatics/PAM_engineering/PAMmla/functions.py", line 178, in predict_from_saved_model
    muts_one_hot = train_test_enc.transform(input_df).toarray()
  File "/home/designer/.local/lib/python3.8/site-packages/sklearn/utils/_set_output.py", line 157, in wrapped
    data_to_wrap = f(self, X, *args, **kwargs)
  File "/home/designer/.local/lib/python3.8/site-packages/sklearn/preprocessing/_encoders.py", line 1027, in transform
    X_int, X_mask = self._transform(
  File "/home/designer/.local/lib/python3.8/site-packages/sklearn/preprocessing/_encoders.py", line 237, in _transform
    self._map_infrequent_categories(X_int, X_mask, ignore_category_indices)
  File "/home/designer/.local/lib/python3.8/site-packages/sklearn/preprocessing/_encoders.py", line 408, in _map_infrequent_categories
    if not self._infrequent_enabled:
AttributeError: 'OneHotEncoder' object has no attribute '_infrequent_enabled'

Thanks in advance.

[justin-delano]

Hello, thank you for trying out the tool! This does seem to be a scikit-learn issue, and we should update our environment to fix this. I will double-check my environment later today, but I seem to remember changing the version of scikit-learn to 1.0.2 worked for me, after removing the package from PAMmla_env.yml and directly adding scikit-learn==1.0.2.

[joanetLiew]

Hi @justin-delano ,
Thanks a lot for your prompt reply.
Switching to sciki-learn 1.0.2 does remove that error. However, one new error popped out.

File "/home/designer/sw/structuralBioinformatics/PAM_engineering/PAMmla/functions.py", line 212, in plot_heatmap
    rate = float(rates[pam].iloc[0])
AttributeError: 'numpy.float64' object has no attribute 'iloc'

I think you could simply change that line to rate = float(rates[pam]) .

Thanks very much again.

[GradientLive]

Hello
I also encountered some problems while trying to install PAMmla.
I want to install PAMmla on a Linux server. When I used conda to create an environment and specified the software version with PAMmla_env.yml, some issues arose that prevented the installation from succeeding. After investigation, I found that the problem may be due to the software version given in the yml file being suitable for Mac servers, not Linux servers.
For example, the given dependencies are: “ - abseil-cpp=20211102.0=hc377ac9_0”

Checking the relevant version information in the conda source will show that the corresponding client is “osx-64”.
Additionally, there are many dependent software packages in the yml file. When installing with conda, resolving the dependency relationship has always failed, resulting in the environment still not being properly configured up to this point.

If possible, could you please provide a Linux dependency environment configuration information? Thank you very much for your team developing such an advanced tool.