mir-am / libtwinsvm Goto Github PK

after installing all the requirements i get this error that says that both numpy library and cython are not found. any idea to fix the issue????

PS C:\Users\Zbook> pip install LIBTwinSVM
Collecting LIBTwinSVM
Using cached LIBTwinSVM-0.3.0.tar.gz (5.2 MB)
Installing build dependencies ... done
Getting requirements to build wheel ... error
error: subprocess-exited-with-error

× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> [31 lines of output]
Traceback (most recent call last):
File "", line 116, in get_numpy_status
ModuleNotFoundError: No module named 'numpy'
Traceback (most recent call last):
File "", line 142, in get_cython_status
ModuleNotFoundError: No module named 'cython'
Traceback (most recent call last):
File "C:\Users\Zbook\AppData\Local\Programs\Python\Python311\Lib\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 353, in
main()
File "C:\Users\Zbook\AppData\Local\Programs\Python\Python311\Lib\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 335, in main
json_out['return_val'] = hook(**hook_input['kwargs'])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Zbook\AppData\Local\Programs\Python\Python311\Lib\site-packages\pip_vendor\pyproject_hooks_in_process_in_process.py", line 118, in get_requires_for_build_wheel
return hook(config_settings)
^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Zbook\AppData\Local\Temp\pip-build-env-8u8l0hbp\overlay\Lib\site-packages\setuptools\build_meta.py", line 341, in get_requires_for_build_wheel
return self._get_build_requires(config_settings, requirements=['wheel'])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Zbook\AppData\Local\Temp\pip-build-env-8u8l0hbp\overlay\Lib\site-packages\setuptools\build_meta.py", line 323, in _get_build_requires
self.run_setup()
File "C:\Users\Zbook\AppData\Local\Temp\pip-build-env-8u8l0hbp\overlay\Lib\site-packages\setuptools\build_meta.py", line 488, in run_setup
self).run_setup(setup_script=setup_script)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Zbook\AppData\Local\Temp\pip-build-env-8u8l0hbp\overlay\Lib\site-packages\setuptools\build_meta.py", line 338, in run_setup
exec(code, locals())
File "", line 322, in
File "", line 256, in setup_package
File "", line 177, in check_install_updated
ImportError: Numerical Python (NumPy) is not installed. LIBTwinSVM requires NumPy >= 1.14.0.
Please check out the installation guide of the LIBTwinSVM:
https://libtwinsvm.readthedocs.io/en/latest/
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error

My code:

from time import process_time
from typing import List, Tuple

import numpy as np
from sklearn.datasets import make_classification
from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split

from libtsvm.estimators import TSVM
from sklearn.svm import SVC as sklearn_SVC


def get_dataset():
    # create synthetic dataset
    X, y = make_classification(n_samples=10000, n_features=100,
                               n_informative=90, n_redundant=10,
                               random_state=0)
    X_train, X_test, y_train, y_test = train_test_split(X, y,
                                                        test_size=.2,
                                                        random_state=0)
    return X_train, X_test, y_train, y_test


def test_twin_SVM(X_train: np.ndarray,
                  X_test: np.ndarray,
                  y_train: np.ndarray,
                  y_test: np.ndarray) \
        -> Tuple[List[float], List[float]]:
    clf = TSVM()
    #clf = sklearn_SVC()
    times = []
    sizes = list(range(1000, 10001, 1000))
    accuracies = []

    print("TwinSVM start")
    for size in sizes:
        print(size)
        X_train_smaller = X_train[:size, :]
        y_train_smaller = y_train[:size]

        time1 = process_time()
        clf.fit(X_train_smaller, y_train_smaller)
        time2 = process_time()
        times.append(time2 - time1)

        acc = accuracy_score(y_test, clf.predict(X_test))
        accuracies.append(acc)
    print()

    return times, accuracies

X_train, X_test, y_train, y_test = get_dataset()
twin_times, twin_accuracies = test_twin_SVM(X_train, X_test,
                                                y_train, y_test)

Output and error:

TwinSVM start
1000

error: index_max(): object has no elements
terminate called after throwing an instance of 'std::logic_error'
  what():  index_max(): object has no elements

What is happening, why am I getting this error? Sklearn SVM is working fine, I've also tried it (commented out in the code above).

from libtsvm.estimators import TSVM
from tensorflow.keras import backend as K
m_temp = tf.keras.Model(inputs=model.input, outputs=model.get_layer('drop_out').output)
t = m_temp.predict(train_ids)
arr = K.eval(t)

lssvc_concat = TSVM(kernel='linear') # Class instantiation

lssvc_concat.fit(arr, train_labels.ravel()) # Fitting the mode

This is my code. I have taken the weights from some deep learning model and I am trying to add the classifier TSVM. Until class instantiation, it is working fine. But while fitting, I am getting like "The kernel appears to have died. It will restart automatically". Kindly help me in this regard.

Thanks in advance...

          I got the same issue. Could anyone guide how to solve this?

Originally posted by @Alexandra97A in #6 (comment)

I am trying to do the TWSVM with OVO. I saw the document and followed the instructions. I got stuck in this part:
'acc, std, full_report = eval_func(h_params)'

When I tried to run the code, I got this error:
'TypeError: Expected sequence or array-like, got <class 'NoneType'>'

I already make sure that X and y are not empty. Can you please help me?