diff options
| -rw-r--r-- | ml_exp/__init__.py | 5 | ||||
| -rw-r--r-- | ml_exp/compound.py | 2 | ||||
| -rw-r--r-- | ml_exp/do_ml.py | 65 | ||||
| -rw-r--r-- | ml_exp/kernels.py | 6 |
4 files changed, 56 insertions, 22 deletions
diff --git a/ml_exp/__init__.py b/ml_exp/__init__.py index c62141a58..48ece56c8 100644 --- a/ml_exp/__init__.py +++ b/ml_exp/__init__.py @@ -25,6 +25,7 @@ from ml_exp.representations import coulomb_matrix, lennard_jones_matrix,\ first_neighbor_matrix, adjacency_matrix, check_bond, bag_of_stuff from ml_exp.math import cholesky_solve from ml_exp.qm7db import qm7db +from ml_exp.do_ml import simple_ml, do_ml __all__ = ['Compound', 'coulomb_matrix', @@ -34,4 +35,6 @@ __all__ = ['Compound', 'check_bond', 'bag_of_stuff', 'cholesky_solve', - 'qm7db'] + 'qm7db', + 'simple_ml', + 'do_ml'] diff --git a/ml_exp/compound.py b/ml_exp/compound.py index 9c7102860..1598fa482 100644 --- a/ml_exp/compound.py +++ b/ml_exp/compound.py @@ -93,8 +93,8 @@ class Compound: bohr_ru=bohr_ru) def gen_am(self, - size=23, use_forces=False, + size=23, bohr_ru=False): """ Generate the Adjacency Matrix for the compund. diff --git a/ml_exp/do_ml.py b/ml_exp/do_ml.py index f32480367..40243d413 100644 --- a/ml_exp/do_ml.py +++ b/ml_exp/do_ml.py @@ -31,19 +31,19 @@ from ml_exp.qm7db import qm7db def simple_ml(descriptors, energies, training_size, - identifier=None, test_size=None, sigma=1000.0, + identifier=None, show_msgs=True): """ Basic ML methodology for a single descriptor type. descriptors: array of descriptors. energies: array of energies. training_size: size of the training set to use. - identifier: string with the name of the descriptor used. test_size: size of the test set to use. If no size is given, the last remaining molecules are used. sigma: depth of the kernel. + identifier: string with the name of the descriptor used. show_msgs: if debug messages should be shown. NOTE: identifier is just a string and is only for identification purposes. Also, training is done with the first part of the data and @@ -86,7 +86,7 @@ def simple_ml(descriptors, mae = np.mean(np.abs(Y_predicted - Y_test)) if show_msgs: - printc('\tMAE for {identifier}: {mae:.4f}', 'GREEN') + printc(f'\tMAE for {identifier}: {mae:.4f}', 'GREEN') toc = time.perf_counter() tictoc = toc - tic @@ -102,20 +102,27 @@ def simple_ml(descriptors, def do_ml(db_path='data', is_shuffled=True, r_seed=111, + diag_value=None, + lj_sigma=1.0, + lj_epsilon=1.0, + use_forces=False, + size=23, + as_eig=True, + bohr_ru=False, + sigma=1000.0, show_msgs=True): """ Main function that does the whole ML process. - training_size: minimum training size. - test_size: size of the test set to use. If no size is given, - the last remaining molecules are used. - ljm_diag_value: if a special diagonal value should be used in lj matrix. - ljm_sigma: sigma value for lj matrix. - ljm_epsilon: epsilon value for lj matrix. + db_path: path to the database directory. + is_shuffled: if the resulting list of compounds should be shuffled. r_seed: random seed to use for the shuffling. - save_benchmarks: if benchmarks should be saved. - size: maximum amount of atoms in molecule. - as_eig: if data should be returned as matrix or array of eigenvalues. - bohr_radius_units: if units should be in bohr's radius units. + diag_value: if special diagonal value is to be used. + lj_sigma: sigma value. + lj_epsilon: epsilon value. + use_forces: if the use of forces instead of k_cx should be used. + size: compound size. + as_eig: if the representation should be as the eigenvalues. + bohr_ru: if radius units should be in bohr's radius units. sigma: depth of the kernel. show_msgs: if debug messages should be shown. """ @@ -134,16 +141,40 @@ def do_ml(db_path='data', # Matrices calculation. tic = time.perf_counter() for compound in compounds: - compound.gen_cm() - compound.gen_ljm() - compound.gen_am() + compound.gen_cm(size=size, + as_eig=as_eig, + bohr_ru=bohr_ru) + compound.gen_ljm(diag_value=diag_value, + sigma=lj_sigma, + epsilon=lj_epsilon, + size=size, + as_eig=as_eig, + bohr_ru=bohr_ru) + compound.gen_am(use_forces=use_forces, + size=size, + bohr_ru=bohr_ru) + + # Create a numpy array for the descriptors. + cm_data = np.array([compound.cm for compound in compounds], dtype=float) + ljm_data = np.array([compound.ljm for compound in compounds], dtype=float) + am_data = np.array([compound.cm for compound in compounds], dtype=float) + print(cm_data.shape, ljm_data.shape, am_data.shape) + toc = time.perf_counter() tictoc = toc - tic if show_msgs: printc(f'Matrices calculation took {tictoc:.4f} seconds.', 'CYAN') # ML calculation. - # PLHLDR + # CM + cm_mae, cm_tictoc = simple_ml(cm_data, + energy_pbe0, + training_size=5000, + test_size=1500, + sigma=1000.0, + identifier='CM', + show_msgs=show_msgs) + print(cm_mae, cm_tictoc) # End of program end_time = time.perf_counter() diff --git a/ml_exp/kernels.py b/ml_exp/kernels.py index 7a61a1e1e..3914ffc20 100644 --- a/ml_exp/kernels.py +++ b/ml_exp/kernels.py @@ -36,9 +36,9 @@ def gaussian_kernel(X1, inv_sigma = -0.5 / (sigma*sigma) K = np.zeros((X1.shape[0], X2.shape[0]), dtype=float) - for i in X1: - for j in X2: - f_norm = np.linalg.norm(i - j) + for i, x1 in enumerate(X1): + for j, x2 in enumerate(X2): + f_norm = np.linalg.norm(x1 - x2) # print(f_norm) K[i, j] = math.exp(inv_sigma * f_norm) |