Trainer

This example shows how to use the Trainer class to train the potential.

import logging

import ase.io
from ase.visualize.plot import plot_atoms

from motep.io.mlip.mtp import read_mtp, write_mtp
from motep.train import Trainer

If you need a log, you can set the following.

logging.basicConfig(level=logging.INFO, format="%(message)s")

We first load the atomic configuration to compute.

images = ase.io.read("../0.0_train/ase.xyz", index=":")
ax = plot_atoms(images[0])

We next load the initial (likely untrained) potential.

mtp_data = read_mtp("initial.mtp")
mtp_data.species = [6, 1]
mtp_data

MTPData(version='1.1.0', potential_name='MTP1m', scaling=1.0, species_count=2, potential_tag='', radial_basis=BasisData(type='RBChebyshev', min=np.float64(0.5), max=np.float64(5.0), size=np.int32(8)), radial_funcs_count=np.int32(1), radial_coeffs=None, alpha_moments_count=np.int32(1), alpha_index_basic_count=np.int32(1), alpha_index_basic=array([[0, 0, 0, 0]], dtype=int32), alpha_index_times_count=np.int32(0), alpha_index_times=array([], shape=(0, 4), dtype=int32), alpha_scalar_moments=np.int32(1), alpha_moment_mapping=array([0], dtype=int32), species_coeffs=None, moment_coeffs=None, _species=array([6, 1], dtype=int32), optimized=['species_coeffs', 'moment_coeffs', 'radial_coeffs'])

We then make the Trainer class and train the potential by the images.

trainer = Trainer(mtp_data, seed=42)
loss = trainer.train(images)

[random seed] = 42
========================================================================

{'method': 'minimize'}

loss 0: 1011.5020145280739
loss 1: 23.966116182125976
loss 2: 16.08095030044797
loss 3: 6.311941371024098
loss 4: 3.161535906684786
loss 5: 0.3927903836126455
loss 6: 0.12596640778521534
loss 7: 0.11544503492176891
loss 8: 0.11307700359591392
loss 9: 0.10964666248733723
loss 10: 0.09822874264626316
loss 11: 0.07973445127246562
loss 12: 0.057272156762290846
loss 13: 0.0566684495683983
loss 14: 0.047560554473403883
loss 15: 0.045952527188951084
loss 16: 0.04546252660298473
loss 17: 0.04446130611485017
loss 18: 0.04148174758171129
loss 19: 0.03539303550150546
loss 20: 0.02559559921709794
loss 21: 0.024678603916939396
loss 22: 0.017362626506152022
loss 23: 0.014945724946532115
loss 24: 0.014683757587025174
loss 25: 0.014673688242865086
loss 26: 0.014648036370317993
loss 27: 0.01457686862424248
loss 28: 0.01388700581382219
loss 29: 0.012861413722844698
loss 30: 0.010702510271064225
loss 31: 0.007803168670421585
loss 32: 0.007441204000437911
loss 33: 0.006749507606295646
loss 34: 0.005139895350261252
loss 35: 0.004919058451915329
loss 36: 0.004834520619320782
loss 37: 0.004832245750094077
loss 38: 0.004829629732815308
loss 39: 0.0048252897556020115
loss 40: 0.004813593974948582
loss 41: 0.004799566962288297
loss 42: 0.004711896386949075
loss 43: 0.00454763233753647
loss 44: 0.0041248465603853915
loss 45: 0.0034454664566298384
loss 46: 0.0030355053594043183
loss 47: 0.002263185488785715
loss 48: 0.0021319881796776413
loss 49: 0.001796806918567014
loss 50: 0.0017746052117819965
loss 51: 0.0017563117646308816
loss 52: 0.0017454259064987512
loss 53: 0.001743781904438849
loss 54: 0.0017425882522220833
loss 55: 0.001738855793639598
loss 56: 0.0017307239585881485
loss 57: 0.0017127256668804096
loss 58: 0.0017030753872952823
loss 59: 0.0016801354882377516
loss 60: 0.0016681680915406492
loss 61: 0.0016521008111056816
loss 62: 0.0016467901590302995
loss 63: 0.0016399961482277134
loss 64: 0.0016340446109405727
loss 65: 0.0016295722506936502
loss 66: 0.0016153413806061592
loss 67: 0.0015953267162001961
loss 68: 0.0015785238847840519
loss 69: 0.0015405535094840642
loss 70: 0.0015073455530816308
loss 71: 0.0014758992190166307
loss 72: 0.0014574073028020993
loss 73: 0.0014445424773020096
loss 74: 0.0014344735382698412
loss 75: 0.0014277498907990568
loss 76: 0.0014195423388780454
loss 77: 0.001415104345783794
loss 78: 0.0014119788711150262
loss 79: 0.001408757962846801
loss 80: 0.001403677962157598
loss 81: 0.0013941441769635002
loss 82: 0.0013889137952071457
loss 83: 0.0013756489779305182
loss 84: 0.0013624252543262695
loss 85: 0.0013457654358195758
loss 86: 0.001329311870590137
loss 87: 0.0013165287552763477
loss 88: 0.0012756635139885004
loss 89: 0.0012619618535257282
loss 90: 0.001260095398214942
loss 91: 0.0012579114550806657
loss 92: 0.0012572842502702655
loss 93: 0.0012564269976886244
loss 94: 0.0012560374121783893
loss 95: 0.0012551315357361556
loss 96: 0.0012545833734606238
loss 97: 0.001254234281429583
loss 98: 0.0012539541980854808
loss 99: 0.001253465227801964
loss 100: 0.001253104408837689
loss 101: 0.001252728443705906
loss 102: 0.001252314470344853
loss 103: 0.0012513011181407752
loss 104: 0.0012503937876135757
loss 105: 0.0012498636001281203
loss 106: 0.0012484013171772887
loss 107: 0.0012476551121246868
loss 108: 0.0012448008406166328
loss 109: 0.0012425793826934285
loss 110: 0.001239549056034909
loss 111: 0.001208166517374574
loss 112: 0.001197677240274903
loss 113: 0.0011737050456611566
loss 114: 0.0011654040341519593
loss 115: 0.0011627113345779197
loss 116: 0.0011608440906147627
loss 117: 0.0011597521173782484
loss 118: 0.0011562279240641504
loss 119: 0.0011558275862683495
loss 120: 0.0011557421385817016
loss 121: 0.001155360226474993
loss 122: 0.0011551759798961118
loss 123: 0.001155049700308418
loss 124: 0.0011548947206116045
loss 125: 0.0011540304710984984
loss 126: 0.0011535103624191433
loss 127: 0.0011518385858093204
loss 128: 0.0011484426379039865
loss 129: 0.00114765872281051
loss 130: 0.0011427761914007558
loss 131: 0.001136973603189545
loss 132: 0.0011334776378167288
loss 133: 0.001131872689226469
loss 134: 0.0011311447479893928
loss 135: 0.0011300885435127504
loss 136: 0.0011294576621975892
loss 137: 0.0011282058132392754
loss 138: 0.0011274355586073643
loss 139: 0.0011259810442767603
loss 140: 0.0011251012626419302
loss 141: 0.001123898582682026
loss 142: 0.0011227319742526565
loss 143: 0.001121339913170174
loss 144: 0.0011198839719247934
loss 145: 0.0011162854286182015
loss 146: 0.0011129583549720967
loss 147: 0.0011084396093107404
loss 148: 0.0011071198978504968
loss 149: 0.0011029115639897386
loss 150: 0.0010951314789772845
loss 151: 0.001092782749729454
loss 152: 0.0010908957641579886
loss 153: 0.0010901006552337285
loss 154: 0.0010871584493972917
loss 155: 0.0010836071837021547
loss 156: 0.0010762198287280723
loss 157: 0.0010611700619032446
loss 158: 0.0010429016918031332
loss 159: 0.0010354117578898097
loss 160: 0.0010228117836357568
loss 161: 0.001009693078962805
loss 162: 0.0010081694171369822
loss 163: 0.0010075722735593253
loss 164: 0.001006046773530871
loss 165: 0.0010039660075021632
loss 166: 0.0010019533997462987
loss 167: 0.0009957451912260846
loss 168: 0.0009916758242831786
loss 169: 0.0009904371856627077
loss 170: 0.0009895261211585772
loss 171: 0.0009878846866147684
loss 172: 0.0009838734409441962
loss 173: 0.0009769247681943479
loss 174: 0.0009678961563686694
loss 175: 0.0009660727981320036
loss 176: 0.0009585804202002342
loss 177: 0.0009554688728581226
loss 178: 0.0009531391493385455
loss 179: 0.0009520221088569601
loss 180: 0.0009515778647903972
loss 181: 0.0009510056791090348
loss 182: 0.0009509091451118447
loss 183: 0.0009502461490118743
loss 184: 0.0009492952676195371
loss 185: 0.0009473234008497291
loss 186: 0.0009453466305249524
loss 187: 0.0009432277712736458
loss 188: 0.0009417231672396018
loss 189: 0.0009405851241451797
loss 190: 0.000940255262322624
loss 191: 0.0009395720777388814
loss 192: 0.0009394845276624465
loss 193: 0.0009392121392631837
loss 194: 0.0009385308667231118
loss 195: 0.0009360626458205651
loss 196: 0.0009337581267871798
loss 197: 0.0009302372123665487
loss 198: 0.000930041310208567
loss 199: 0.0009287546662549961
loss 200: 0.0009277430525512186
loss 201: 0.0009272045359567569
loss 202: 0.0009260544469566486
loss 203: 0.0009251788306971644
loss 204: 0.0009240733514400784
loss 205: 0.0009233365129838973
loss 206: 0.000922826544614842
loss 207: 0.0009225882760398221
loss 208: 0.0009215008071521195
loss 209: 0.0009210982216749375
loss 210: 0.0009190882876684012
loss 211: 0.0009154889943272209
loss 212: 0.0009109180850349601
loss 213: 0.0009045315017510923
loss 214: 0.000904436554413071
loss 215: 0.0009042732450238724
loss 216: 0.0009042641477146275
loss 217: 0.000904190422994636
loss 218: 0.0009040645169467096
loss 219: 0.0009038533732149971
loss 220: 0.0009037541829460332
loss 221: 0.0009034569572324156
loss 222: 0.0009033004889691364
loss 223: 0.0009032187306936391
loss 224: 0.0009030646814976727
loss 225: 0.0009028938733766278
loss 226: 0.0009025008877517109
loss 227: 0.0009018679336636983
loss 228: 0.0009011381818931397
loss 229: 0.0009010619447570565
loss 230: 0.0009006014296333162
loss 231: 0.0009004105243034601
loss 232: 0.0009003068197812133
loss 233: 0.000900128419978605
loss 234: 0.0008997540817960647
loss 235: 0.0008992608864637561
loss 236: 0.0008990157268746579
loss 237: 0.0008988228606956432
loss 238: 0.0008987180586051638
loss 239: 0.000898641011443072
loss 240: 0.0008984297935652109
loss 241: 0.0008980377720344907
loss 242: 0.0008973667296272916
loss 243: 0.0008972730848934304
loss 244: 0.0008964457549658707
loss 245: 0.0008952626626715276
loss 246: 0.0008932567079042049
loss 247: 0.0008908192769483782
loss 248: 0.0008883594772319014
loss 249: 0.0008871971082615381
loss 250: 0.0008863561425566058
loss 251: 0.000886223358479388
loss 252: 0.000886130725135833
loss 253: 0.000885807350564387
loss 254: 0.0008855380961482706
loss 255: 0.0008854105181592767
loss 256: 0.0008853369204769254
loss 257: 0.0008852504196349071
loss 258: 0.000885040065826625
loss 259: 0.0008846522999451692
loss 260: 0.0008836378020691892
loss 261: 0.0008824496710105294
loss 262: 0.0008813606908208217
loss 263: 0.0008802761670456251
loss 264: 0.0008796997619424704
loss 265: 0.0008795970326422446
loss 266: 0.0008794705073614292
loss 267: 0.0008790252796403996
loss 268: 0.0008789385973568817
loss 269: 0.000878803041226661
loss 270: 0.0008786494948659127
loss 271: 0.0008782301240505049
loss 272: 0.000877615401715192
loss 273: 0.0008773744794251173
loss 274: 0.000876739816977392
loss 275: 0.0008761829525465402
loss 276: 0.0008755517771766904
loss 277: 0.0008753286779549426
loss 278: 0.0008749210629818658
loss 279: 0.0008745614788027427
loss 280: 0.0008741796881725263
loss 281: 0.000873437791925954
loss 282: 0.0008720702956307973
loss 283: 0.0008698960119170347
loss 284: 0.0008694454103224168
loss 285: 0.0008679185083865806
loss 286: 0.0008666561427256717
loss 287: 0.0008665079988342759
loss 288: 0.0008664414691107298
loss 289: 0.0008662281778732992
loss 290: 0.0008661366701670609
loss 291: 0.0008657182008645651
loss 292: 0.000865332197294431
loss 293: 0.0008650018587599356
loss 294: 0.0008645462953607889
loss 295: 0.0008642381426505964
loss 296: 0.000863698445141591
loss 297: 0.0008633607742064783
loss 298: 0.0008622431803662279
loss 299: 0.0008612210508527126
loss 300: 0.0008589933407250411
loss 301: 0.0008573168818157817
loss 302: 0.000855512600586147
loss 303: 0.0008536301558725063
loss 304: 0.0008533276590999168
loss 305: 0.0008524514481463414
loss 306: 0.0008523179982356647
loss 307: 0.0008519912538026928
loss 308: 0.0008517314121627206
loss 309: 0.0008512015175242997
loss 310: 0.0008503913396231429
loss 311: 0.0008489285068092635
loss 312: 0.0008476171959955148
loss 313: 0.0008469919185572019
loss 314: 0.0008462992398122177
loss 315: 0.0008422818100601417
loss 316: 0.0008373558185227549
loss 317: 0.0008328642214931736
loss 318: 0.0008312020733034382
loss 319: 0.0008273262033841497
loss 320: 0.0008234850264342833
loss 321: 0.0008195982647246386
loss 322: 0.0008165041610302248
loss 323: 0.0008084183361996371
loss 324: 0.000800136584220867
loss 325: 0.0007951705506916586
loss 326: 0.0007942502605492041
loss 327: 0.0007934785424319438
loss 328: 0.0007928647607953856
loss 329: 0.0007924060651621976
loss 330: 0.0007916287991185253
loss 331: 0.000790921053656726
loss 332: 0.0007906348553627373
loss 333: 0.0007880960208839748
loss 334: 0.000784556927377578
loss 335: 0.0007838892780516381
loss 336: 0.0007780992824627386
loss 337: 0.0007723910190470265
loss 338: 0.0007699475660889371
loss 339: 0.000769637292259393
loss 340: 0.0007691785078051967
loss 341: 0.0007688176809499462
loss 342: 0.0007685956414287518
loss 343: 0.0007682001381229885
loss 344: 0.0007674977300770554
loss 345: 0.0007669228882423998
loss 346: 0.000766080358708658
loss 347: 0.0007647091192637584
loss 348: 0.0007642717816131446
loss 349: 0.0007633621605821
loss 350: 0.0007625378618107257
loss 351: 0.0007621558249941476
loss 352: 0.000761544818277621
loss 353: 0.0007608023125399881
loss 354: 0.0007591158366159786
loss 355: 0.0007549593856588812
loss 356: 0.0007523945125009306
loss 357: 0.0007487090870624569
loss 358: 0.0007464564044576844
loss 359: 0.000745342556592875
loss 360: 0.0007445787659558067
loss 361: 0.0007440046970700855
loss 362: 0.0007430931387394592
loss 363: 0.0007424127208639096
loss 364: 0.0007404804717591503
loss 365: 0.0007402620016447444
loss 366: 0.0007394457631310824
loss 367: 0.0007380809169444348
loss 368: 0.0007375465201068878
loss 369: 0.0007368352053006738
loss 370: 0.000736499157567128
loss 371: 0.000735198472350854
loss 372: 0.0007327725464452009
loss 373: 0.0007269921192268251
loss 374: 0.0007252758903299434
loss 375: 0.0007185606639388077
loss 376: 0.0006939036774062298
loss 377: 0.000662302269837773
loss 378: 0.0006602855787084945
loss 379: 0.0006380586435249868
loss 380: 0.0006278253113631124
loss 381: 0.0006267353251332982
loss 382: 0.0006253520767108412
loss 383: 0.0006240922641914466
loss 384: 0.0006215290060802909
loss 385: 0.0006179651130670613
loss 386: 0.0006174451371387645
loss 387: 0.0006156496859855984
loss 388: 0.0006150978301003358
loss 389: 0.0006147658285321546
loss 390: 0.0006142299115410117
loss 391: 0.0006139332139408724
loss 392: 0.0006131354654476284
loss 393: 0.0006114101120715401
loss 394: 0.0006106139448787548
loss 395: 0.0006088504906837922
loss 396: 0.0006033584441844432
loss 397: 0.0005973671021749816
loss 398: 0.0005962903682696861
loss 399: 0.0005939489873102485
loss 400: 0.000588416561045302
loss 401: 0.0005819314442285021
loss 402: 0.0005742477280282858
loss 403: 0.0005724721453939196
loss 404: 0.0005620393780553129
loss 405: 0.0005590576029500735
loss 406: 0.0005503531057078077
loss 407: 0.0005428029021247609
loss 408: 0.0005365205322608444
loss 409: 0.0005312049017636508
loss 410: 0.0005297392071769119
loss 411: 0.0005260154868234651
loss 412: 0.0005258030447158984
loss 413: 0.0005255302242061189
loss 414: 0.0005251668731712129
loss 415: 0.000525009306679184
loss 416: 0.0005243461095486929
loss 417: 0.0005239143759832899
loss 418: 0.0005234994265285946
loss 419: 0.0005229574294462172
loss 420: 0.0005216747892076214
loss 421: 0.000519841905450137
loss 422: 0.0005195865217158786
loss 423: 0.0005180264423191388
loss 424: 0.0005161796876296032
loss 425: 0.0005151109135745071
loss 426: 0.0005142218460677635
loss 427: 0.0005129980559885736
loss 428: 0.000509926708251652
loss 429: 0.000500036051860955
loss 430: 0.0004898074932183576
loss 431: 0.00048380878745681646
loss 432: 0.00048324223267414736
loss 433: 0.0004825503354237535
loss 434: 0.0004825080056525429
loss 435: 0.00048245009775325004
loss 436: 0.00048238598651000313
loss 437: 0.00048230158405510585
loss 438: 0.00048209626066585316
loss 439: 0.00048205875052744705
loss 440: 0.00048179226850093675
loss 441: 0.00048164353490812644
loss 442: 0.0004813904603306892
loss 443: 0.0004811602284025922
loss 444: 0.0004811047903595644
loss 445: 0.0004810531919696664
loss 446: 0.00048100469107185376
loss 447: 0.00048099260254804474
loss 448: 0.0004809842936437481
loss 449: 0.0004809448688119955
loss 450: 0.0004808775485959855
loss 451: 0.0004807041988295077
loss 452: 0.0004805622840195917
loss 453: 0.00048028821744029086
loss 454: 0.000479724393552439
loss 455: 0.00047909417681754184
loss 456: 0.00047827646278955307
loss 457: 0.0004772799915760314
loss 458: 0.0004744956009830034
loss 459: 0.00047433707793185267
loss 460: 0.0004738787558598422
loss 461: 0.00047378450222800397
loss 462: 0.0004737280011776414
loss 463: 0.0004735511923681209
loss 464: 0.0004735081136395678
loss 465: 0.00047346831098457625
loss 466: 0.0004733115248227415
loss 467: 0.00047324577660322933
loss 468: 0.00047314894896640215
loss 469: 0.0004730618068103319
loss 470: 0.00047303875916423864
loss 471: 0.000472960583626403
loss 472: 0.0004727465302760088
loss 473: 0.0004724170417768771
loss 474: 0.00047233047301922044
loss 475: 0.0004722418707583394
loss 476: 0.0004720564213549469
loss 477: 0.00047200542682895137
loss 478: 0.00047197727881141216
loss 479: 0.0004719443528553991
loss 480: 0.00047192207962621973
loss 481: 0.0004718743250930109
loss 482: 0.00047181363841116987
loss 483: 0.00047180215477544925
loss 484: 0.00047178327829330364
loss 485: 0.00047173074100452365
loss 486: 0.00047163603464183687
loss 487: 0.00047155254269170774
loss 488: 0.0004714228415889644
loss 489: 0.0004713275344896815
loss 490: 0.00047121709233249176
loss 491: 0.00047117768414975224
loss 492: 0.0004710713704713336
loss 493: 0.00047103308669171846
loss 494: 0.00047101328920968094
loss 495: 0.0004709973222863973
loss 496: 0.0004709839374252073
loss 497: 0.00047097979763995556
loss 498: 0.00047097601087496506
loss 499: 0.00047096565953669815
loss 500: 0.0004709469428037056
loss 501: 0.00047091506751587454
loss 502: 0.0004708806479922126
loss 503: 0.0004708336083373
loss 504: 0.00047076196510167164
loss 505: 0.0004707165413976261
loss 506: 0.00047068633674493803
loss 507: 0.0004706433778986131
loss 508: 0.00047058629901383176
loss 509: 0.0004705434923805234
loss 510: 0.0004705229658887787
loss 511: 0.00047050851002667835
loss 512: 0.00047049564375711123
loss 513: 0.0004704876202003454
loss 514: 0.00047048317088053634
loss 515: 0.00047047150559437174
loss 516: 0.00047046325066681723
loss 517: 0.00047045309694444164
loss 518: 0.00047044680755188107
loss 519: 0.00047044170652324994
loss 520: 0.00047039819722562983
loss 521: 0.0004703040469372194
loss 522: 0.00047023834811560117
loss 523: 0.00047021628673223775
loss 524: 0.0004700878348694996
loss 525: 0.00046976692319308897
loss 526: 0.0004693894009332566
loss 527: 0.000469201816778186
loss 528: 0.0004689152955290345
loss 529: 0.00046859883749663295
loss 530: 0.0004685318494579072
loss 531: 0.0004684203545016306
loss 532: 0.00046835058453027854
loss 533: 0.0004682833715688853
loss 534: 0.00046811505898829777
loss 535: 0.0004679683251263922
loss 536: 0.00046786610769003115
loss 537: 0.00046781713778035433
loss 538: 0.0004677865907599015
loss 539: 0.0004677476081278521
loss 540: 0.00046768634856055095
loss 541: 0.00046756559039874235
loss 542: 0.00046754655561907354
loss 543: 0.0004674252232150013
loss 544: 0.00046733100585993567
loss 545: 0.0004672599014827242
loss 546: 0.0004671737423581755
loss 547: 0.00046704655037638535
loss 548: 0.0004670286338509681
loss 549: 0.00046689863572733215
loss 550: 0.0004668183479317106
loss 551: 0.0004667696722902021
loss 552: 0.00046670021638212924
loss 553: 0.00046654902470855
loss 554: 0.0004664425344684395
loss 555: 0.00046634798979173825
loss 556: 0.00046630360836587254
loss 557: 0.00046629253365197465
loss 558: 0.00046625631551558915
loss 559: 0.00046606973841119444
loss 560: 0.00046603903067767905
loss 561: 0.00046593940780430537
loss 562: 0.0004658789220857991
loss 563: 0.0004658549962300774
loss 564: 0.0004658196242999079
loss 565: 0.0004657563823451525
loss 566: 0.0004656957842452573
loss 567: 0.0004656021559088843
loss 568: 0.000465498459281498
loss 569: 0.00046545923608742553
loss 570: 0.00046540260225566213
loss 571: 0.0004653960565389088
loss 572: 0.00046537592130750156
loss 573: 0.0004653565713219574
loss 574: 0.00046534541492235763
loss 575: 0.00046533591282184246
loss 576: 0.0004653265759465281
loss 577: 0.0004653215367488406
loss 578: 0.00046530640071258043
loss 579: 0.00046528519750919033
loss 580: 0.00046523000293731696
loss 581: 0.0004650718485069272
loss 582: 0.0004647107445708454
loss 583: 0.0004645159512383963
loss 584: 0.0004639783921494713
loss 585: 0.0004634058483272725
loss 586: 0.0004632302540673061
loss 587: 0.000463200167239848
loss 588: 0.000463139775503945
loss 589: 0.0004630464919619743
loss 590: 0.00046290936028622187
loss 591: 0.000462874837365225
loss 592: 0.0004628164676586527
loss 593: 0.00046276203069906327
loss 594: 0.0004626983652509445
loss 595: 0.00046268366583760486
loss 596: 0.000462639661397219
loss 597: 0.0004626082767320629
loss 598: 0.0004626004658930469
loss 599: 0.0004625765648595476
loss 600: 0.0004625590778630798
loss 601: 0.0004625341765202649
loss 602: 0.0004625212319471406
loss 603: 0.0004624988293254042
loss 604: 0.0004624464434118952
loss 605: 0.00046237625112147124
loss 606: 0.00046224889318790684
loss 607: 0.0004622337000603921
loss 608: 0.0004621989098945609
loss 609: 0.00046216327247223014
loss 610: 0.00046215468177707733
loss 611: 0.0004620856688738053
loss 612: 0.0004620241978359175
loss 613: 0.00046197155205487386
loss 614: 0.00046194342852241924
loss 615: 0.00046190920377249506
loss 616: 0.000461881465158464
loss 617: 0.0004618485213654842
loss 618: 0.00046179911143027656
loss 619: 0.0004617720022428753
loss 620: 0.0004617387806494585
loss 621: 0.0004616753009482455
loss 622: 0.00046160862946570565
loss 623: 0.0004615461209327742
loss 624: 0.00046140034228624833
loss 625: 0.0004612978570692698
loss 626: 0.0004611275389973027
loss 627: 0.00046097381349333866
loss 628: 0.0004608907765087688
loss 629: 0.00046083020512099356
loss 630: 0.00046081392634276527
loss 631: 0.00046076932989961184
loss 632: 0.00046070714014487366
loss 633: 0.00046069177610334836
loss 634: 0.0004605988278755519
loss 635: 0.0004605410770786334
loss 636: 0.00046048942807848956
loss 637: 0.0004604560379379951
loss 638: 0.00046043133329517924
loss 639: 0.00046042223059179375
loss 640: 0.00046041469027465544
loss 641: 0.00046040288541962743
loss 642: 0.0004603864081612034
loss 643: 0.00046035414888270825
loss 644: 0.0004603273021068349
loss 645: 0.00046028415869046917
loss 646: 0.0004602783316827851
loss 647: 0.0004602559513687851
loss 648: 0.00046024657620620296
loss 649: 0.00046024286309511646
loss 650: 0.0004602383634602932
loss 651: 0.0004602309502493512
loss 652: 0.0004602075012024856
loss 653: 0.0004601622045756613
loss 654: 0.00046008638711271895
loss 655: 0.0004600682131730922
loss 656: 0.0004600052190008921
loss 657: 0.00045998104295150833
loss 658: 0.00045997471966297644
loss 659: 0.000459966857122207
loss 660: 0.00045995392812680705
loss 661: 0.0004599302488056848
loss 662: 0.0004598980664193247
loss 663: 0.00045987399745001627
loss 664: 0.0004598330144061726
loss 665: 0.0004597953672891622
loss 666: 0.0004597766779113487
loss 667: 0.0004597705667127578
loss 668: 0.0004597650562883103
loss 669: 0.0004597464551490651
loss 670: 0.00045973670885836006
loss 671: 0.0004597197928462835
loss 672: 0.0004596956798748167
loss 673: 0.00045968511027084203
loss 674: 0.0004596588810816484
loss 675: 0.00045963575409776396
loss 676: 0.00045960480968234993
loss 677: 0.0004595433668966473
loss 678: 0.0004595274772657228
loss 679: 0.00045942748714825873
loss 680: 0.0004593485599553374
loss 681: 0.0004592761985900492
loss 682: 0.0004592187641824041
loss 683: 0.0004591922668109626
loss 684: 0.00045912082778830414
loss 685: 0.00045907975654985617
loss 686: 0.00045906085130115545
loss 687: 0.0004590540797359458
loss 688: 0.00045904475825114626
loss 689: 0.00045903813703405743
loss 690: 0.0004590301490697871
loss 691: 0.0004590164970230511
loss 692: 0.0004589993653507892
loss 693: 0.00045897886734211693
loss 694: 0.0004589504083501155
loss 695: 0.00045889206997172175
loss 696: 0.00045887178774602425
loss 697: 0.00045878410322247587
loss 698: 0.00045865160847346337
loss 699: 0.0004585599633069099
loss 700: 0.0004584705025058353
loss 701: 0.00045839776375224497
loss 702: 0.0004583775585185953
loss 703: 0.00045835285429329987
loss 704: 0.0004583261513759865
loss 705: 0.00045829601854993813
loss 706: 0.00045824728674517985
loss 707: 0.000458211979536338
loss 708: 0.00045816666330582566
loss 709: 0.00045814919294420275
loss 710: 0.00045814277974064453
loss 711: 0.0004581392392704945
loss 712: 0.00045812985393840083
loss 713: 0.0004581264452268926
loss 714: 0.00045811907376004624
loss 715: 0.00045811490359153246
loss 716: 0.00045810710827916046
loss 717: 0.000458097004935524
loss 718: 0.0004580653259425251
loss 719: 0.0004580636145955965

Optimization result:
  Message: CONVERGENCE: RELATIVE REDUCTION OF F <= FACTR*EPSMCH
  Success: True
  Status code: 0
  Number of function evaluations: 824
  Number of iterations: 719

Energy (eV):
    Errors checked for 101 configurations
    MAX error: 0.055490966571511535
    ABS error: 0.010121549203893998
    RMS error: 0.013690262970662826

Energy per atom (eV/atom):
    Errors checked for 101 configurations
    MAX error: 0.006936370821438942
    ABS error: 0.0012651936504867497
    RMS error: 0.0017112828713328533

Forces per component (eV/angstrom):
    Errors checked for 808 atoms
    MAX error: 0.608851736536841
    ABS error: 0.09332206272079513
    RMS error: 0.12317130693061584

Stress per component (GPa):
    Errors checked for 0 configurations
    MAX error: nan
    ABS error: nan
    RMS error: nan

Time (step 0: minimize): 40.040620960999995 (s)

We can see the final loss function value.

loss(mtp_data.parameters)

np.float64(0.0004580636145955965)

After the training, the given mtp_data is updated in-place. We finally store the trained potential in a new file.

write_mtp("final.mtp", mtp_data)