molSimplify 1.7.4__py3-none-any.whl

molSimplify/python_krr/y_train_TS.csv

@@ -0,0 +1,535 @@
+0.8279554328413462
+-1.5894446888418896
+-0.8136741825830084
+1.4190657286412627
+0.5399386347006085
+1.4292794521809697
+-1.5418972678888163
+-1.5406755213016197
+0.5375590476448919
+-1.9123678725269404
+-0.3585372726514036
+-0.027458542816885388
+1.4867608498742315
+-1.458900088139217
+0.7330567286633817
+-1.1512384835859442
+-0.5184480566007958
+0.7822100102191492
+-1.3374922551265578
+-0.995322672227324
+-0.14399429658640311
+-0.10403455527164561
+-0.20576915451973782
+0.39658123608783424
+-1.4855589551259976
+2.0089298091982646
+0.5268091302732347
+0.8942384486275566
+-0.7692548496106856
+1.9902804008727464
+-1.630715290361875
+-1.6581694059839553
+2.2593843845366917
+0.8816985265762344
+-0.28370209248896205
+0.5290172691358842
+0.883876605678588
+-0.8413429095816979
+-1.3149128439242002
+-1.0027817377437214
+0.4197423172111754
+1.2242366860527683
+-0.11653236084265259
+-0.546491862713397
+-0.1762604807499548
+0.5419234407467245
+1.8806498204348647
+-1.5401340279276357
+-0.994000992365876
+0.9864919799586147
+-0.021573194165399334
+-1.076514284850125
+-1.1642685907195773
+0.1872543787302102
+-1.5397335809486712
+-0.08397760148131353
+-1.5367387202448906
+0.5205270409872192
+0.5308785114439418
+0.5481700503212702
+0.5243791083304203
+0.33418543479804247
+-1.558712595772578
+-0.9943033306906371
+-0.9877788840213417
+-1.1354711383194862
+0.868086377580053
+-0.5294332102231252
+-1.24042271683141
+0.5133864709731073
+-0.5824054425466456
+0.5256742246122627
+-1.5423527313385104
+0.5248193918343774
+-1.4635787032272485
+-0.06492145962497117
+0.5359709443096291
+0.5334720201944824
+-1.555299843997115
+0.631908789845134
+0.34837603960538144
+-1.320674369051341
+-1.0995610129087103
+1.1371044128004404
+-0.307634358481302
+2.037073522560279
+0.5351864484706992
+-0.9960426750158359
+-1.5618335357515003
+0.09178630464784719
+1.3057758695054094
+-0.5803827517018845
+-0.7947945768274632
+-0.2214821415378758
+-0.9940955468720094
+0.5441414423699782
+0.5329971930080327
+-0.4162940593866578
+-0.08224765700169077
+0.9080723368379018
+0.5261248687659523
+0.5425853512911885
+-1.5431001499840076
+2.016859303348087
+0.5274250806399993
+1.022987616568101
+-1.5556124508866394
+0.5408130109168521
+0.5329428026591647
+0.5294995907265143
+0.7316116493011293
+1.7169778997354699
+1.4105321165058888
+1.2307955462821811
+0.5991520148267626
+0.5307412461873834
+-1.5458146948750993
+1.5197431947386093
+0.5401851405624912
+-1.4820734140512841
+1.9928924666831644
+-0.34120951451501236
+0.5409018693962843
+-1.3728496303117168
+2.3327972840217233
+0.5765468807770255
+1.4148222160354142
+0.894820036261204
+0.5206291000035926
+0.028559244763715862
+0.5369490991557699
+0.6273591955919974
+0.9091584079492561
+-1.56298263541976
+0.5335630791709176
+-0.32939529865123107
+0.46748658778469954
+-0.7623887366487144
+-1.028629804734449
+0.4566944775649624
+0.13757824065593327
+0.5258042432842207
+0.6069315936017738
+0.5259504084641489
+-1.262048628380868
+-0.035128025241171326
+-1.5561695146476018
+-0.42204368527534947
+0.5217712589214706
+0.6192276661415739
+0.2149007012708594
+-0.29175828278976845
+1.826175067624409
+0.887734066449486
+-1.5428878223750924
+-0.07077448124154888
+-0.06015465289594516
+-0.7799846957771287
+-1.083422482625313
+0.6150573745258147
+0.5385214666791195
+-1.4200801212241851
+-0.993903040562331
+-0.373256035506832
+0.5120687468851223
+0.8780597501532075
+-0.2473442729753254
+-0.7864485186559121
+2.0089057440356677
+0.8620718725508993
+-1.6744733745624942
+-0.14275904998402822
+0.8771120461287911
+0.05495511415941851
+0.38024583549710356
+-0.515824148718251
+-0.6485080126387303
+0.15403584747743232
+-0.7727214690391003
+1.852469276174475
+0.5386028989514431
+-0.16790848543535278
+-0.9918589789464565
+-1.5330066049572442
+0.3831257201465158
+-0.15138908391435582
+0.17032529738458782
+1.873119291840861
+0.5457032583645615
+0.8877557052579155
+1.9933655087564066
+-1.4915182723121263
+0.5386563456225558
+-1.0551864814909646
+0.08120679039034374
+-0.1636213988624678
+-1.5381214864004946
+-1.4902548324814904
+0.5354453210841611
+-1.4584070748539955
+0.5264972035522774
+-1.5436193755365573
+-0.2202343214103847
+-0.6432770925418109
+-0.6794307413692833
+-0.9395826665106211
+-1.434205120351477
+0.4639893277248605
+0.4932143512790398
+0.36049614951776177
+0.5381753327043868
+0.5320692759046153
+0.0128735522162463
+0.7811841072202289
+2.0165280663896863
+-0.08533408000367854
+0.6115586186819622
+-1.018230577106256
+-0.3399873521169751
+1.5856459477528528
+-0.8274116671657429
+-1.692908450867656
+0.5886938707540138
+1.4827000126404282
+2.048224363056447
+-1.2046492172285674
+0.8735362746455324
+-0.4484987009962453
+-0.24880842329170158
+0.856275743362906
+0.7157369078955317
+0.07665768706704044
+-0.04621979350949226
+-0.06299367747076944
+-0.11984539125008138
+0.8819603541400227
+0.5265131733683858
+0.739311365428582
+-1.3607476209906941
+0.44251516348544023
+-0.892928440590135
+-1.0628961637413974
+0.5384371938898127
+-0.6656628918909374
+-2.198531532582757
+-0.5051553139078702
+0.5325591913640982
+-0.4429013362861379
+2.319908873287969
+-0.569569081360624
+-0.060848551981723314
+-0.5808724199683113
+-0.5675642992270141
+-0.2569521069316974
+-0.6783919547961611
+0.8313168964468071
+-0.5568238375671156
+-1.3049726563336106
+0.38690786281749956
+1.235186072532382
+-0.9942336911964021
+0.054820652087550006
+-1.537013667285674
+1.6866377738130856
+-1.0954517597695554
+1.2225716306084335
+-1.9574875663373434
+-0.5350490257426536
+-0.07221208514946219
+-0.1026470123474035
+1.8672528405450275
+0.17254276981143762
+0.14539739585675499
+-1.4141132912808803
+0.0247411013587314
+-0.5602451967538469
+-0.07105867738307896
+-0.6425298195980215
+-0.20650967947914423
+-0.06384756996439422
+0.39098250957855973
+-1.1816854053833517
+-1.5584005643391117
+0.5303581358112558
+-1.5429068602242915
+-1.5413552000203974
+-1.4276029940364299
+-0.7210655385460952
+-0.032802592616742725
+-0.9925255814081246
+0.5269367905018665
+-1.4735428518314524
+0.8041639343797155
+0.5394721726628694
+1.7361475299402283
+0.5380501211766425
+-0.6941912477475496
+1.5234686190720848
+-0.9891614178784071
+0.8716158559807949
+0.5448062207739869
+0.9137277894837077
+0.38730573893878545
+-0.24074105486232172
+-0.9954566338372216
+0.8803343092667433
+-0.53267728174408
+0.5694415311422002
+-1.0782838143500741
+0.5874779807082308
+-0.45935702171794923
+-1.450554478199492
+0.536309147980943
+-0.16822195340606094
+-0.0738034597057069
+-1.4601645590106367
+-0.2560460003472999
+0.530151812628342
+2.039155375344389
+-1.2621905877864101
+0.5365300311389052
+-0.9969832656925199
+-0.0647093238491315
+-0.5246188113550154
+2.412838550673534
+1.4114849030319028
+0.485877360454777
+0.5315924298776747
+0.5359541604976966
+0.5370100613347819
+0.5271258325481923
+0.4930179809957355
+-1.070283945396139
+2.0313890287241185
+0.8726091291605103
+0.5379405087181774
+1.4209895801715375
+-0.018021530451420156
+-0.9923359073184431
+-0.5083547661570859
+-0.5271932235088296
+1.8868275577635616
+-0.685101214255827
+-1.0200555338106188
+0.8822442448024794
+-0.5838730802065201
+0.5330540925384952
+-1.5349905154333605
+-1.5122422254381254
+0.5457506847209822
+0.7391612800616542
+0.5341359961697948
+0.5374110189122427
+-1.1059234508412386
+2.0037735956656166
+1.6413550352916904
+0.4901111771179128
+-1.1487154867260754
+-1.1419848821247813
+-0.1845063795986699
+-0.1798907862431564
+-0.602059965499556
+-1.1436586892903122
+-0.8724475557570035
+-0.8809226150334999
+1.7129453752699406
+-1.0229674472277561
+0.6287965134484941
+0.3582842675271805
+-0.48128608215613466
+-1.834689091437175
+-1.366942248182176
+-1.4999685955154576
+0.13124524002371243
+-0.9967455041740843
+-0.6452541096435969
+-0.6683699326888941
+0.4764320194736921
+-0.5471724927083494
+-0.6530808057063496
+0.06714419671018462
+1.2131740534197069
+0.9559138360504031
+0.5366286901880568
+2.0104845802440536
+-0.5206142471462498
+2.3279489628974614
+0.5318145155454076
+1.993128631053573
+-0.9938375068669798
+0.5396143021002999
+-0.167494035384514
+-0.4976204436466864
+0.8759648561647757
+-0.36846234836368763
+-1.3459857687034098
+-1.3945146935211823
+-1.3698616190369144
+-0.3342109128447633
+0.9474489607802992
+-1.4423758950394028
+0.531618612404315
+0.5187198993988903
+-0.5642811140708539
+0.5382069592972937
+-1.0862389843830964
+0.4732936481143914
+0.20867019019039149
+-0.07809407295090127
+-1.163168797719365
+-1.3079917369974443
+-0.41295546424719626
+-0.4275350945503871
+-0.04957898797331479
+0.5304493085194987
+0.830846285252735
+-0.49246342205163873
+-0.5153332642564555
+1.2510143857478433
+0.5271371029677518
+0.5381750963852737
+0.5354647282027196
+0.5325261721836504
+0.5246584996268319
+0.5374959560427904
+-0.7102969105897917
+2.6203501856249356
+-1.541357072402849
+1.9969488916341438
+-0.07905193099471974
+1.2223396296009448
+-0.9874857846062511
+-0.5006431461941571
+0.47654559029479654
+0.5398543299062133
+0.22192398231268304
+-0.9955067946836519
+1.2252587319180321
+-0.7714525760284542
+0.12458390286161249
+0.46856740715925804
+-0.5474251894224224
+0.529526611493952
+0.008017415662944952
+0.30194899001674785
+-0.7865920863182115
+1.2783450380605816
+0.8739871339843436
+0.8900890820969937
+-1.2412391595657677
+-0.06707850610796774
+0.4367829191083737
+-1.4094302575687747
+1.8483956711079332
+0.5357533649431336
+-0.6380355717897958
+-0.6564641140792372
+-1.3823659497531324
+1.2640033964514448
+-1.1403844108106513
+-1.015104985089211
+-0.7983810180852915
+-1.6477652622739305
+-0.5811066495863813
+0.5359488288035693
+-0.8282062166299334
+0.9700070996201163
+1.7133349308946162
+2.3352096406037823
+0.5214784425705856
+0.4837736117705238
+-0.7014271863614058
+1.2359297243197749
+-0.42974300232681917
+0.5280133931376895
+-1.0663771782289346
+0.01863531999612446
+-0.5449764136280935
+1.5250965422356944
+0.5421825493635813
+0.28959581811408325
+0.22982366875750646
+0.5323298941167536
+0.5336079372065736
+1.7209825771423728
+-0.38914424768012196
+-0.29317165144416096
+-1.5541973201190862
+-1.4536405160617094
+0.5362826522784125
+0.3549422267128743
+0.24494908995638603
+-1.0074604090239925
+-0.5248789882212362
+0.5833184901402053
+-0.6170941868255954
+1.8715841991175435
+-0.990326368447058
+2.165597482012762
+-1.812632177721212
+0.5307815113843117
+-0.6407542165067629
+-1.273415951088272
+-1.5416552484862476
+1.416359000984037
+0.5278557286961949
+1.7237887067130364
+-1.139098504282394
+-0.07245723152644076
+0.04203720230007345
+-0.03499658875650629
+0.813193107562947
+-0.5017025851965705
+1.4971161490111824
+-0.07722541885061691
+-0.12612734759562325
+0.46904637824844386
+-0.269936758428369
+0.6673226302314803
+0.245186502170593
+-0.5247215029173177
+1.711181280235031
+-0.391984952856525
+-0.9964066797456341
+0.5289790014583213
+-0.9910588648552312
+0.5345743712841656
+-0.8966788791928417
+0.536058133721006
+0.5405730478115114
+0.5484158482015611
+-0.9955405340922531
+0.6963177856471717
+-0.5706754277225492
+-0.6938570529723639

molSimplify/python_nn/ANN.py

@@ -0,0 +1,198 @@
+# Written by JP Janet for HJK Group
+# Dpt of Chemical Engineering, MIT
+
+# ########################################################
+# ###### This script contains a neural network  ##########
+# ###  trained on octahedral metal-ligand          #######
+# ######   bond distances and spin propensity  ###########
+# ########################################################
+
+
+import csv
+
+import numpy as np
+from importlib_resources import files as resource_files
+from molSimplify.utils.decorators import deprecated
+from typing import List
+
+
+def simple_network_builder(layers: List[int], partial_path: str):
+    """Numpy based implementation of a simple neural network to replace the
+    now deprecated pybrain variant."""
+
+    class ThreeLayerNetwork():
+        """Fixed architecture neural network"""
+
+        def __init__(self, layers: List[int], partial_path: str):
+            self.w1 = np.array(
+                csv_loader(partial_path + '_w1.csv')).reshape(-1, layers[0])
+            self.w2 = np.array(
+                csv_loader(partial_path + '_w2.csv')).reshape(-1, layers[1])
+            self.w3 = np.array(
+                csv_loader(partial_path + '_w3.csv')).reshape(-1, layers[2])
+            self.b1 = np.array(csv_loader(partial_path + '_b1.csv'))
+            self.b2 = np.array(csv_loader(partial_path + '_b2.csv'))
+            self.b3 = np.array(csv_loader(partial_path + '_b3.csv'))
+
+        def activate(self, input: np.ndarray) -> np.ndarray:
+            layer1 = np.tanh(self.w1 @ input + self.b1)
+            layer2 = np.tanh(self.w2 @ layer1 + self.b2)
+            output = self.w3 @ layer2 + self.b3
+            return output
+
+    return ThreeLayerNetwork(layers, partial_path)
+
+
+@deprecated
+def simple_network_builder_pybrain(layers: List[int], partial_path: str):
+    from pybrain.structure import (FeedForwardNetwork, TanhLayer, LinearLayer,
+                                   BiasUnit, FullConnection)
+    n = FeedForwardNetwork()
+    # create the network
+    inlayer = LinearLayer(layers[0], name="In")
+    hidden_one = TanhLayer(layers[1], name="Hidden 1")
+    hidden_two = TanhLayer(layers[2], name="Hidden 2")
+    b1 = BiasUnit(name="Bias")
+    output = LinearLayer(1, name="Out")
+    n.addInputModule(inlayer)
+    n.addModule(hidden_one)
+    n.addModule(hidden_two)
+    n.addModule(b1)
+    n.addOutputModule(output)
+    in_to_one = FullConnection(inlayer, hidden_one)
+    one_to_two = FullConnection(hidden_one, hidden_two)
+    two_to_out = FullConnection(hidden_two, output)
+    b1_to_one = FullConnection(b1, hidden_one)
+    b2_to_two = FullConnection(b1, hidden_two)
+    b3_to_output = FullConnection(b1, output)
+    # load weights and biases
+    in_to_one._setParameters(np.array((csv_loader(partial_path + '_w1.csv'))))
+    one_to_two._setParameters(np.array(csv_loader(partial_path + '_w2.csv')))
+    two_to_out._setParameters(np.array(csv_loader(partial_path + '_w3.csv')))
+    b1_to_one._setParameters(np.array(csv_loader(partial_path + '_b1.csv')))
+    b2_to_two._setParameters(np.array(csv_loader(partial_path + '_b2.csv')))
+    b3_to_output._setParameters(np.array(csv_loader(partial_path + '_b3.csv')))
+
+    # connect the network topology
+    n.addConnection(in_to_one)
+    n.addConnection(one_to_two)
+    n.addConnection(two_to_out)
+    # n.sortModules()
+
+    n.addConnection(b1_to_one)
+    n.addConnection(b2_to_two)
+    n.addConnection(b3_to_output)
+
+    # finalize network object
+    n.sortModules()
+
+    return n
+
+
+def csv_loader(path: str) -> List[float]:
+    # print('in csv loader')
+    path_to_file = resource_files("molSimplify.python_nn").joinpath(path.strip("/"))
+    with open(path_to_file, 'r') as csvfile:
+        csv_lines = csv.reader(csvfile, delimiter=',')
+        ret_list = list()
+        for lines in csv_lines:
+            this_line = [float(a) for a in lines]
+            ret_list += this_line
+    return ret_list
+
+
+def matrix_loader(path, rownames=False):
+    # loads matrix with rowname option
+    path_to_file = resource_files("molSimplify.python_nn").joinpath(path.strip("/"))
+    if rownames:
+        with open(path_to_file, "r") as f:
+            csv_lines = list(csv.reader(f))
+            row_names = [row[0] for row in csv_lines]
+            mat = [row[1:] for row in csv_lines]
+        return mat, row_names
+    else:
+        with open(path_to_file, 'r') as csvfile:
+            csv_lines = csv.reader(csvfile, delimiter=',')
+            mat = [a for a in csv_lines]
+        return mat
+
+
+# n = network_builder([25,50,51],"nn_split")
+def simple_splitting_ann(excitation):
+    # print('path to ANN data: ',path_to_file)
+    n = simple_network_builder([25, 50, 50], "ms_split")
+    excitation, sp_center, sp_shift = excitation_standardizer(excitation, 'split')
+    # print(excitation)
+    # print('center is ' + str(sp_center))
+    # print('scale is '+ str(sp_shift))
+    # print(excitation)
+    result = n.activate(excitation)
+
+    # print('result is ' + str(result))
+    result = (result*sp_shift) + sp_center
+    # print('result is ' + str(result))
+    return result, excitation
+
+
+def simple_slope_ann(slope_excitation):
+    # print('path to ANN data: ',path_to_file)
+    n = simple_network_builder([24, 50, 50], "ms_slope")  # no alpha value
+    # print(slope_excitation)
+    slope_excitation, sl_center, sl_shift = excitation_standardizer(slope_excitation, 'slope')
+    # print(slope_excitation)
+    result = n.activate(slope_excitation)
+    # print('result is ' + str(result))
+    # print('center is ' + str(sl_center) + ' shift  '+ str(sl_shift))
+    result = (result*sl_shift) + sl_center
+    # print('result is ' + str(result))
+    return result
+
+
+def simple_ls_ann(excitation):
+    n = simple_network_builder([25, 50, 50], "ms_ls")
+    excitation, ls_center, ls_shift = excitation_standardizer(excitation, 'ls')
+    result = n.activate(excitation)
+    result = result*ls_shift + ls_center
+    return result
+
+
+def simple_hs_ann(excitation):
+    n = simple_network_builder([25, 50, 50], "ms_hs")
+    excitation, hs_center, hs_shift = excitation_standardizer(excitation, 'hs')
+    result = n.activate(excitation)
+    result = result*hs_shift + hs_center
+    return result
+
+
+def excitation_standardizer(excitation, tag):
+    """This function implements a scale-and-center type of normalization
+    that may help predictions currently testing for splitting and slope only
+    """
+
+    centers = csv_loader(tag+"_center.csv")
+    shifts = csv_loader(tag+"_scale.csv")
+    descriptor_centers = np.array(centers[1:])
+    descriptor_shifts = np.array(shifts[1:])
+    sp_center = centers[0]
+    sp_shift = shifts[0]
+    excitation = np.array(excitation)
+    excitation = (excitation - descriptor_centers)
+    excitation = np.divide(excitation, descriptor_shifts)
+    return(excitation, sp_center, sp_shift)
+
+
+def find_eu_dist(excitation):
+    # returns euclidean distance to nearest trainning
+    # vector in desciptor space
+    mat, rownames = matrix_loader('train_data.csv', rownames=True)
+    train_mat = np.array(mat, dtype='float64')
+    min_dist = 1000
+    excitation, _, _ = excitation_standardizer(excitation, 'split')
+    for i, rows in enumerate(train_mat):
+        np.subtract(rows, np.array(excitation))
+        this_dist = np.linalg.norm(np.subtract(rows, np.array(excitation)))/3
+        if this_dist < min_dist:
+            min_dist = this_dist
+            best_row = rownames[i]
+    # print('min dist is ' +str(min_dist))
+    return min_dist, best_row