fastgplearn.backend package¶
Submodules¶
fastgplearn.backend.p_numpy module¶
- fastgplearn.backend.p_numpy.cos_(a, b)¶
- fastgplearn.backend.p_numpy.exp_(a, b)¶
- fastgplearn.backend.p_numpy.find_add_mask(popi, single_start=6)¶
- fastgplearn.backend.p_numpy.find_add_mask_all(pop, single_start=6)¶
- fastgplearn.backend.p_numpy.find_used_index(popi, single_start=6)¶
- fastgplearn.backend.p_numpy.find_used_index_total(pop, single_start=6)¶
- fastgplearn.backend.p_numpy.get_corr_together(fake_ys, y)¶
- Parameters
fake_ys (np.ndarray) – with shape (n_results, n_sample,).
y (np.ndarray) – with sample (n_sample,).
- Returns
with shape (n_result,)
- Return type
corr (np.ndarray)
- fastgplearn.backend.p_numpy.get_sort_accuracy_together(fake_ys, y)¶
- Parameters
fake_ys (np.ndarray) – with shape (n_results, n_sample,).
y (np.ndarray) – with sample (n_sample,).
- Returns
with shape (n_result,)
- Return type
corr (np.ndarray)
- fastgplearn.backend.p_numpy.ln_(a, b)¶
- fastgplearn.backend.p_numpy.max_(a, b)¶
- fastgplearn.backend.p_numpy.min_(a, b)¶
- fastgplearn.backend.p_numpy.p_np_cal(ve, xs, y, func_index=None, single_start=6)¶
Batch calculate.
- fastgplearn.backend.p_numpy.p_np_score(ve, xs, y, func_index, clf=False, single_start=6)¶
Batch score.
- fastgplearn.backend.p_numpy.p_np_score_mp(ve, xs, y, func_index=None, n_jobs=1, clf=False, single_start=6)¶
Batch score with n_jobs.
- fastgplearn.backend.p_numpy.p_np_str_name(ve, xns, cns=None, y=None, func_index=None, real_names=None)¶
Batch get name of expression,(without coef and intercept).
- fastgplearn.backend.p_numpy.pow2_(a, b)¶
- fastgplearn.backend.p_numpy.pow3_(a, b)¶
- fastgplearn.backend.p_numpy.rec_(a, b)¶
- fastgplearn.backend.p_numpy.sin_(a, b)¶
fastgplearn.backend.p_torch module¶
- fastgplearn.backend.p_torch.cos_(a, b)¶
- fastgplearn.backend.p_torch.exp_(a, b)¶
- fastgplearn.backend.p_torch.get_corr_together(fake_ys, y)¶
- Parameters
fake_ys (torch.Tensor) – with shape (n_results, n_sample,).
y (torch.Tensor) – with sample (n_sample,).
- Returns
with shape (n_result,)
- Return type
corr (torch.Tensor)
- fastgplearn.backend.p_torch.get_sort_accuracy_together(fake_ys, y)¶
- Parameters
fake_ys (torch.ndarray) – with shape (n_results, n_sample,).
y (torch.ndarray) – with sample (n_sample,).
- Returns
with shape (n_result,)
- Return type
corr (torch.ndarray)
- fastgplearn.backend.p_torch.ln_(a, b)¶
- fastgplearn.backend.p_torch.max_(a, b)¶
- fastgplearn.backend.p_torch.min_(a, b)¶
- fastgplearn.backend.p_torch.p_torch_cal(ve, xs, y, func_index=None, single_start=6)¶
Batch calculate.
- fastgplearn.backend.p_torch.p_torch_score(ve, xs, y, func_index, return_numpy=False, clf=False, single_start=6)¶
Batch score.
- fastgplearn.backend.p_torch.p_torch_score_mp(ve, xs, y, func_index=None, n_jobs=1, return_numpy=False, clf=False, single_start=6)¶
Batch score with n_jobs. It’s slow!!!
- fastgplearn.backend.p_torch.pow2_(a, b)¶
- fastgplearn.backend.p_torch.pow3_(a, b)¶
- fastgplearn.backend.p_torch.rec_(a, b)¶
- fastgplearn.backend.p_torch.sin_(a, b)¶