for d in diff5:
EVOM_error += d * d
print("Difference between Compounds by representation:")
print("BOB", BOB_error)
print("CM", CM_error)
print("EVCM", EVCM_error)
print("OM", OM_error)
print("EVOM", EVOM_error)
if do_derivative_calculation:
#create results instances
mols, compounds = datprep.read_xyz_energies(database)
datprep.store_compounds(compounds, database + "compounds.pickle")
#prepare derivatives of all representations
replist = [ZRNrep.Coulomb_Matrix, ZRNrep.Eigenvalue_Coulomb_Matrix, ZRNrep.Bag_of_Bonds, ZRNrep.Overlap_Matrix, \
ZRNrep.Eigenvalue_Overlap_Matrix]
for i in range(5):
results, resultaddition = jader.calculate_num_der(
replist[i], compounds)
res_file = database + namelist[i] + "der_results.pickle"
datprep.store_compounds(results, res_file)
print("results were successfully stored to ", res_file)
if do_plot_derivatives:
yvals = []
norms_nuc = []
###analytical derivative:
#results = jader.calculate_eigenvalues('CM_EV', compound_ls)
###numerical derivative:
results, resultaddition = jader.calculate_num_der(numerical_representations[2],
compound_ls)
###B)
###store list of results in result_file
'''
#If you want to plot from multiple pickle results file, use this code:
#result_file = result_folder + "results_%i-%i.pickle" %(init, end)
'''
print("results:", results)
print("len of results:", len(results))
datprep.store_compounds(results, results_file_OM)
#C)
#If you want to plot from multiple pickle results file, use this code:
#these are used as file identifiers for the results_%i-%i.pickle files
#numbers = ["0-200", "200-400", "400-600", "600-800", "800-1000", "1000-1200", "1200-1400", "1400-1600", "1600-1800", "1800-2000", "2000-2200", "2200-2400", "2400-2600", "2600-2800", "2800-3000", "3000-3200", "3200-3400", "3400-3600", "3600-3800", "3800-3993"]
'''
#read list of compounds from data file
full_compound_ls = datprep.read_compounds(results_file)
#print(len(full_compound_ls), " compounds in full data file")
'''
###use if only part of dataset should be processed
#try:
# compound_ls = full_compound_ls[init : end]
K_test = m_c.laplacian_kernel_matrix(
x_training=m_c.x_training, x_test=m_c.x_test)
m_c.test_predicted_results = np.dot(K_test, alphas)
mae = m_c.calculate_mae()
mae_nmodels += mae
print("mae_nmodels ", mae_nmodels)
avg_mae = mae_nmodels / float(nModels)
print("avg mae:", avg_mae)
m_c.mae = avg_mae
learning_list.append(m_c)
mae_list.append(avg_mae)
print("totally tested:", total_tested)
name = rep_names[
rep] #"sigma: %.2e, lambda: %.2e" %(sigma, lamda)
curve = datprep.CurveObj(name)
curve.xnparray = training_no
curve.ynparray = np.array(mae_list)
curve_list.append(curve)
final_file = "./tmp/Kernel_Results/" + final_file_list[rep]
datprep.store_compounds(curve_list, final_curve_file)
compounds = datprep.read_compounds(small_data_file)
single = False
if single:
for c in compounds:
ev, vectors = jrep.CM_ev(c.Z, c.R, c.N)
print("name of compound:", c.filename)
#print("eigenvalue repro:\n", ev)
derivative = jader.sort_derivative('CM_EV', c.Z, c.R, c.N, 2, "R", "R")
print(derivative)
results, fractions = jader.calculate_eigenvalues('CM_EV', compounds)
datprep.store_compounds(results, CM_ev_result_file)
print(type(results[0]))
#y-axis information
dZ_percentages = []
dR_percentages = []
dZdZ_percentages = []
dRdR_percentages = []
dZdR_percentages = []
#x-axis information
norms = []
#C)
#get all the data from our results list
def full_kernel_ridge(fingerprint_list,
property_list,
result_file,
set_sizes,
sigmas=[],
lambdas=[],
rep_no=1,
upperlimit=12,
Choose_Folder=False,
representation="CM"):
#print("result_file:", result_file)
''' Kernel ridge regression model
y(X') = sum_i alpha_i K(X', X_i)
Input
-----
fingerprint_list : list of fingerprints
property_list : list of learning data, e.g. energy values corresponding to the fingerprints
result_file : file where data is stored with pickle.
training_size : desired size of training set
sigmas : fitting coefficient
lambdas : fitting coefficient
upperlimit : int, total of training + test set. Can be used if more data is available than is being used or to bootstrap
Choose_Folder: boolean, if True, file is directly stored to result_file.
if not, result file is stored in ./Pickled/Kernel_Results folder
representation: str, abbreviation for fingerprint used
Return
------
learning_list : list of LearningResults Objects
raw_data_files : list of names of files where raw data was stored to
Stored
------
raw data is stored to raw_data_file entries, learning_list is sotred to result_file
'''
start = tic()
learning_list = []
raw_data_files = []
if not Choose_Folder:
print("your results are stored to ./Pickled/Kernel_Results/")
result_file = "./Pickled/Kernel_Results/" + result_file + "_" + str(
rep_no) + "reps"
#loop over learning defined by number of repetition, sigmas, and lamdas
for i in range(rep_no):
for s in sigmas:
for l in lambdas:
#for every i, s, l combination, a new Learning Object is created and stored to the learning list
maes = []
for sets in set_sizes:
t1 = tic()
#make training and test list:
training_indices, test_indices = make_training_test(
len(fingerprint_list), sets, upperlim=upperlimit)
#print("training:", training_indices)
#print("test:", test_indices)
tr_fingerprints = [
fingerprint_list[i] for i in training_indices
]
tr_properties = [
property_list[i] for i in training_indices
]
tr_size = len(training_indices)
tst_fingerprints = [
fingerprint_list[i] for i in test_indices
]
tst_properties = [property_list[i] for i in test_indices]
t2 = tic()
K = build_kernel_matrix(tr_fingerprints, tr_size, s)
t3 = tic()
#print("\n \n \nkernel matrix:\n ", K)
#get alpha coefficients
alphas = get_alphas(K, tr_properties, tr_size, l)
t4 = tic()
#print("\n \n \n alphas:\n ", alphas)
#print("trainin/test split:", t2 - t1)
#print("kernel matrix:", t3-t2)
#print("alphas calculation:", t4 - t3)
#predict properties of test set
results, errors = predict_new(s, alphas, tr_fingerprints,
tst_fingerprints,
tst_properties)
mae = sum(abs(errors)) / (len(errors))
maes.append(mae)
#save raw data
filename = './tmp/%srawdata_rep%i_sigma%s_lamda%f_set%i.dat' % (
representation, i, str(s), l, sets)
raw_data_files.append(filename)
save_raw_data(filename, tr_properties, training_indices,
tst_properties, results, test_indices)
#add learning result to list
learning_list.append(
LearningResults(l, s, np.array(set_sizes), np.array(maes)))
print("round %i successfully finished" % (i + 1))
#save maes with data so it can be plotted
datprep.store_compounds(learning_list, result_file)
return (learning_list, raw_data_files)
with this file XYZ files can be converted to database_preparation.compounds class objects
'''
#define path to folder containing xyz files
database = "../Databases/QM9_XYZ/"
#define path to where you want to store your data
database_file = "../Databases/Pickled/qm9.pickle"
#define path to where you want to store data of molecules with
#less heavy atoms than in the database_file
dat_ha_file = "../Databases/Pickled/qm7.pickle"
#read all compounds in database file and convert to datprep class objects
mol_ls, compound_ls = datprep.read_xyz_energies(database)
#store compounds to database_file
datprep.store_compounds(compound_ls, database_file)
#store all compounds with less than 7 atoms with code below:
datprep.sortby_heavyatoms(database_file, dat_ha_file, 7)
ha_compounds = datprep.read_compounds(dat_ha_file)
for c in compound_ls:
print(c.heavy_atoms())
print("now all heavy atoms in sorted list")
for i in ha_compounds:
print(i.heavy_atoms())
if rep == 1:
M = ZRN_rep.Eigenvalue_Coulomb_Matrix_h(compound.Z, compound.R)
mol.representation = M.flatten()
if rep == 2:
mol.generate_bob(asize={'C': 7, 'H': 16, 'N': 6, 'O': 4, 'F': 4})
if rep == 3:
M = ZRN_rep.Overlap_Matrix_h(compound.Z, compound.R)
mol.representation = M.flatten()
if rep == 4:
M = ZRN_rep.Eigenvalue_Overlap_Matrix_h(compound.Z, compound.R)
mol.representation = M.flatten()
#add representation array to X_list
X_list[rep].append(mol.representation)
#prepare Kernel_Result raw instances (no training/test split, no sigma, no lamda)
m = datprep.Kernel_Result()
m.representation_name = rep_names[rep]
m.x = X_list[rep]
m.y = Y_energy_list
CM_list.append(m)
datprep.store_compounds(CM_list,
"./tmp/%s_raw_Kernel_Results" % rep_names[rep])
