update for keras 3.0

kgcnn/utils/plots.py

@@ -37,11 +37,15 @@ def plot_train_test_loss(histories: list, loss_name: str = None,
         loss_name = [x for x in list(histories[0].keys()) if "val_" not in x]
     if val_loss_name is None:
         val_loss_name = [x for x in list(histories[0].keys()) if "val_" in x]
-
     if not isinstance(loss_name, list):
         loss_name = [loss_name]
     if not isinstance(val_loss_name, list):
         val_loss_name = [val_loss_name]
+    if not isinstance(data_unit, list):
+        data_unit = [data_unit]
+
+    if len(data_unit) < len(val_loss_name):
+        data_unit = data_unit + [str(data_unit[-1])]*(len(val_loss_name)-len(data_unit))
 
     train_loss = []
     for x in loss_name:
@@ -74,7 +78,7 @@ def plot_train_test_loss(histories: list, loss_name: str = None,
         plt.scatter([len(train_loss[i][0])], [np.mean(y, axis=0)[-1]],
                     label=r"{0}: {1:0.4f} $\pm$ {2:0.4f} ".format(
                         val_loss_name[i], np.mean(y, axis=0)[-1],
-                        np.std(y, axis=0)[-1]) + data_unit, color=vp[0].get_color()
+                        np.std(y, axis=0)[-1]) + data_unit[i], color=vp[0].get_color()
                     )
     plt.xlabel('Epochs')
     plt.ylabel('Loss')

training/results/QM9Dataset/Schnet/Schnet_QM9Dataset_score_U.yaml

@@ -0,0 +1,270 @@
+OS: nt_win32
+backend: tensorflow
+cuda_available: 'False'
+data_unit: '[''eV'']'
+date_time: '2023-10-28 15:35:41'
+device_id: '[LogicalDevice(name=''/device:CPU:0'', device_type=''CPU'')]'
+device_memory: '[]'
+device_name: '[{}]'
+epochs:
+- 800
+- 800
+- 800
+- 800
+- 800
+- 800
+- 800
+- 800
+- 800
+- 800
+execute_folds: []
+kgcnn_version: 4.0.0
+learning_rate:
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+loss:
+- 0.003102271119132638
+- 0.0031446837820112705
+- 0.003222967963665724
+- 0.0031532456632703543
+- 0.003179531078785658
+- 0.003303215140476823
+- 0.0032826880924403667
+- 0.0033122405875474215
+- 0.003219034057110548
+- 0.0031931130215525627
+max_learning_rate:
+- 0.0005000000237487257
+- 0.0005000000237487257
+- 0.0005000000237487257
+- 0.0005000000237487257
+- 0.0005000000237487257
+- 0.0005000000237487257
+- 0.0005000000237487257
+- 0.0005000000237487257
+- 0.0005000000237487257
+- 0.0005000000237487257
+max_loss:
+- 0.32900306582450867
+- 0.31977978348731995
+- 0.315213680267334
+- 0.3267098367214203
+- 0.32087260484695435
+- 0.32263708114624023
+- 0.33119043707847595
+- 0.32553577423095703
+- 0.33107835054397583
+- 0.3280688524246216
+max_scaled_mean_absolute_error:
+- 0.32905193320264625
+- 0.3198247700749311
+- 0.3152591766524487
+- 0.3267146232519014
+- 0.32090585330008337
+- 0.32266499042006824
+- 0.331219953791748
+- 0.3255783607024572
+- 0.33108641226527835
+- 0.3280781928206874
+max_scaled_root_mean_squared_error:
+- 0.48803940748218383
+- 0.4722164986112966
+- 0.46687197334126235
+- 0.4824642231699354
+- 0.47237326096568216
+- 0.4767886644757702
+- 0.48718154684191073
+- 0.48111225911341654
+- 0.4896235333899374
+- 0.48458047356738754
+max_val_loss:
+- 0.13950803875923157
+- 0.09146694093942642
+- 0.08519700914621353
+- 0.09302295744419098
+- 0.07964996248483658
+- 0.09005503356456757
+- 0.08714553713798523
+- 0.08930784463882446
+- 0.14616762101650238
+- 0.08222486078739166
+max_val_scaled_mean_absolute_error:
+- 0.1384251979366305
+- 0.09089954509881863
+- 0.08493482909156742
+- 0.09195206884545477
+- 0.07788716696188017
+- 0.08892871728155785
+- 0.08499441338207311
+- 0.08826035654976387
+- 0.14532977491327478
+- 0.08147701913348938
+max_val_scaled_root_mean_squared_error:
+- 0.18370449513677514
+- 0.13094710846372856
+- 0.1282176691272395
+- 0.13305046672874024
+- 0.13000283246683073
+- 0.13205004523360545
+- 0.13780788845402625
+- 0.127828771209062
+- 0.1859635789862445
+- 0.12361408291501223
+min_learning_rate:
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+- 1.06999996205559e-05
+min_loss:
+- 0.003102271119132638
+- 0.0031446837820112705
+- 0.003222967963665724
+- 0.0031532456632703543
+- 0.003179531078785658
+- 0.003303215140476823
+- 0.0032826880924403667
+- 0.0033122405875474215
+- 0.003219034057110548
+- 0.0031931130215525627
+min_scaled_mean_absolute_error:
+- 0.00310240541088055
+- 0.003145102704026026
+- 0.003223467196040231
+- 0.0031533548883815768
+- 0.0031800479318645027
+- 0.0033036366300158255
+- 0.0032831822678082764
+- 0.003312745464602815
+- 0.0032194061774542034
+- 0.0031933249947576877
+min_scaled_root_mean_squared_error:
+- 0.004267732501010393
+- 0.004246839030432776
+- 0.004458374770604839
+- 0.004275395080879121
+- 0.00431145122420274
+- 0.005204495085082492
+- 0.004462491839127707
+- 0.0046641875080630545
+- 0.0045305375165805035
+- 0.004296213755401796
+min_val_loss:
+- 0.014781356789171696
+- 0.012920898385345936
+- 0.01292756199836731
+- 0.014434972777962685
+- 0.014712795615196228
+- 0.013618988916277885
+- 0.014511722140014172
+- 0.014128431677818298
+- 0.012978708371520042
+- 0.013119565322995186
+min_val_scaled_mean_absolute_error:
+- 0.014120716750712173
+- 0.012860281101251151
+- 0.012784171533398017
+- 0.013679683258432147
+- 0.013684824596960582
+- 0.0129447451102775
+- 0.013640340616333098
+- 0.013602984113398517
+- 0.012948668216790582
+- 0.012709200020117863
+min_val_scaled_root_mean_squared_error:
+- 0.049919464270817504
+- 0.03026642499533745
+- 0.03513453907510313
+- 0.04346468396620625
+- 0.039105222914479355
+- 0.035136985111746795
+- 0.04161567543048673
+- 0.03471576091403692
+- 0.02727225442509016
+- 0.03069353773982778
+model_class: make_model
+model_name: Schnet
+model_version: ''
+multi_target_indices:
+- 11
+number_histories: 10
+scaled_mean_absolute_error:
+- 0.00310240541088055
+- 0.003145102704026026
+- 0.003223467196040231
+- 0.0031533548883815768
+- 0.0031800479318645027
+- 0.0033036366300158255
+- 0.0032831822678082764
+- 0.003312745464602815
+- 0.0032194061774542034
+- 0.0031933249947576877
+scaled_root_mean_squared_error:
+- 0.004267732501010393
+- 0.004246839030432776
+- 0.004458374770604839
+- 0.004275395080879121
+- 0.00431145122420274
+- 0.005204495085082492
+- 0.004462491839127707
+- 0.0046641875080630545
+- 0.0045305375165805035
+- 0.004296213755401796
+seed: 42
+time_list:
+- '7:59:50.696123'
+- '7:48:22.017206'
+- '7:55:55.432746'
+- '7:49:30.672315'
+- '7:55:53.938183'
+- '10:09:28.508289'
+- '8:52:12.519615'
+- '8:38:43.179948'
+- '8:09:18.160746'
+- '8:25:30.579937'
+val_loss:
+- 0.015042244456708431
+- 0.012920898385345936
+- 0.01292756199836731
+- 0.014434972777962685
+- 0.014712795615196228
+- 0.013618988916277885
+- 0.014511722140014172
+- 0.014128431677818298
+- 0.013102271594107151
+- 0.013119565322995186
+val_scaled_mean_absolute_error:
+- 0.014382796675197521
+- 0.012860281101251151
+- 0.012784171533398017
+- 0.013679683258432147
+- 0.013684824596960582
+- 0.0129447451102775
+- 0.013640340616333098
+- 0.013602984113398517
+- 0.013072252945093014
+- 0.012709200020117863
+val_scaled_root_mean_squared_error:
+- 0.050914259932424766
+- 0.030465935479548335
+- 0.03528550774068547
+- 0.0436440285448667
+- 0.039105222914479355
+- 0.035136985111746795
+- 0.04161567543048673
+- 0.03471576091403692
+- 0.027616748351973794
+- 0.03166351491351372

training/results/QM9Dataset/Schnet/Schnet_hyper_U.json

@@ -0,0 +1 @@
+{"model": {"class_name": "make_model", "module_name": "kgcnn.literature.Schnet", "config": {"name": "Schnet", "inputs": [{"shape": [null], "name": "node_number", "dtype": "int32"}, {"shape": [null, 3], "name": "node_coordinates", "dtype": "float32"}, {"shape": [null, 2], "name": "range_indices", "dtype": "int64"}, {"shape": [], "name": "total_nodes", "dtype": "int64"}, {"shape": [], "name": "total_ranges", "dtype": "int64"}], "cast_disjoint_kwargs": {"padded_disjoint": false}, "input_node_embedding": {"input_dim": 95, "output_dim": 64}, "last_mlp": {"use_bias": [true, true, true], "units": [128, 64, 1], "activation": ["kgcnn>shifted_softplus", "kgcnn>shifted_softplus", "linear"]}, "interaction_args": {"units": 128, "use_bias": true, "activation": "kgcnn>shifted_softplus", "cfconv_pool": "scatter_sum"}, "node_pooling_args": {"pooling_method": "scatter_sum"}, "depth": 4, "gauss_args": {"bins": 20, "distance": 4, "offset": 0.0, "sigma": 0.4}, "verbose": 10, "output_embedding": "graph", "use_output_mlp": false, "output_mlp": null, "output_scaling": {"name": "ExtensiveMolecularLabelScaler"}}}, "training": {"fit": {"batch_size": 32, "epochs": 800, "validation_freq": 10, "verbose": 2, "callbacks": [{"class_name": "kgcnn>LinearLearningRateScheduler", "config": {"learning_rate_start": 0.0005, "learning_rate_stop": 1e-05, "epo_min": 100, "epo": 800, "verbose": 0}}]}, "compile": {"optimizer": {"class_name": "Adam", "config": {"learning_rate": 0.0005}}, "loss": {"class_name": "kgcnn>MeanAbsoluteError", "config": {"dtype": "float64"}}, "metrics": [{"class_name": "MeanAbsoluteError", "config": {"dtype": "float64", "name": "scaled_mean_absolute_error"}}, {"class_name": "RootMeanSquaredError", "config": {"dtype": "float64", "name": "scaled_root_mean_squared_error"}}]}, "scaler": {"class_name": "ExtensiveMolecularLabelScaler", "config": {"atomic_number": "node_number"}}, "multi_target_indices": [11]}, "data": {}, "info": {"postfix": "", "postfix_file": "_U", "kgcnn_version": "4.0.0"}, "dataset": {"class_name": "QM9Dataset", "module_name": "kgcnn.data.datasets.QM9Dataset", "config": {}, "methods": [{"set_train_test_indices_k_fold": {"n_splits": 10, "random_state": 42, "shuffle": true}}, {"map_list": {"method": "set_range", "max_distance": 4, "max_neighbours": 30}}, {"map_list": {"method": "count_nodes_and_edges", "total_edges": "total_ranges", "count_edges": "range_indices", "count_nodes": "node_number", "total_nodes": "total_nodes"}}]}}

training/train_graph.py

@@ -96,7 +96,7 @@
     train_indices_all.append(train_index)
 
     # Only do execute_splits out of the k-folds of cross-validation.
-    if execute_folds:
+    if execute_folds is not None:
         if current_split not in execute_folds:
             continue
     print("Running training on split: '%s'." % current_split)