Training in progress, step 10000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c0489f4d8a0feab5d3e38c38764efb3065a72c0df1166c52788a4fb8b5db167
+oid sha256:9663ba3ab5044bc90a7b3d1b75f640ff55f2b5c41c418701a4ae83c867c824bb
 size 409608164

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:10a21a5a2f88c53aa60c7d7a14adc335d7340e642ca9d190920e4e42bed1705e
+oid sha256:72cc2c8c672b182b4bcc420e9df023d1155dae42a4b06a811b042e577bd5ae25
 size 814647162

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:12f5435a54f71ddfe7351c79962b19289d71b582fb8054a3e03ea18859c1e069
+oid sha256:e527a24b032ccf5b3ad099345582f9a572dd9ca9f0a87b20d04145f903aac542
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7c3ad366a455d36d790316ad7de46df5551299c0b50500acfaef90eebc7bbcf
+oid sha256:8d92c5a2febec6a3debc54e07f631908af44ee341af8b138c2ac3072c6cfbde8
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": 0.0,
   "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_cosine-opt25/trial-number=21-learning_rate=1.6e-06-warmup_ratio=0.080-num_cycles=1.500/checkpoint-5000",
-  "epoch": 0.49319392385085814,
+  "epoch": 0.9863878477017163,
   "eval_steps": 2500,
-  "global_step": 5000,
+  "global_step": 10000,
   "is_hyper_param_search": true,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -119,6 +119,160 @@
       "eval_tanimoto_similarity": 0.0,
       "eval_valid": 0.14653115333627928,
       "step": 5000
+    },
+    {
+      "epoch": 0.542513316235944,
+      "grad_norm": 1.294830083847046,
+      "learning_rate": 9.304780137412219e-07,
+      "loss": 0.8837,
+      "step": 5500
+    },
+    {
+      "epoch": 0.5918327086210298,
+      "grad_norm": 1.3187594413757324,
+      "learning_rate": 1.302486007417367e-06,
+      "loss": 0.8177,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6411521010061156,
+      "grad_norm": 1.3196697235107422,
+      "learning_rate": 1.5501941814704978e-06,
+      "loss": 0.739,
+      "step": 6500
+    },
+    {
+      "epoch": 0.6904714933912014,
+      "grad_norm": 1.3168965578079224,
+      "learning_rate": 1.6113648928799203e-06,
+      "loss": 0.6635,
+      "step": 7000
+    },
+    {
+      "epoch": 0.7397908857762873,
+      "grad_norm": 1.1013790369033813,
+      "learning_rate": 1.4706287622758396e-06,
+      "loss": 0.598,
+      "step": 7500
+    },
+    {
+      "epoch": 0.7397908857762873,
+      "eval_all_ligands_equal": 0.0,
+      "eval_e3_equal": 0.06637207247017234,
+      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
+      "eval_e3_graph_edit_distance_norm": 1.0,
+      "eval_e3_has_attachment_point(s)": 0.9723376049491825,
+      "eval_e3_heavy_atoms_difference": 0.5682722050375607,
+      "eval_e3_heavy_atoms_difference_norm": -0.02346762473633578,
+      "eval_e3_tanimoto_similarity": 0.0,
+      "eval_e3_valid": 0.9723376049491825,
+      "eval_has_all_attachment_points": 0.7742819266460451,
+      "eval_has_three_substructures": 0.9734865223155104,
+      "eval_heavy_atoms_difference": 61.96756517896598,
+      "eval_heavy_atoms_difference_norm": 0.8021929226054672,
+      "eval_linker_equal": 0.0015024304021210783,
+      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
+      "eval_linker_graph_edit_distance_norm": 1.0,
+      "eval_linker_has_attachment_point(s)": 0.943791427308882,
+      "eval_linker_heavy_atoms_difference": 8.018294299602298,
+      "eval_linker_heavy_atoms_difference_norm": 0.38053119346826625,
+      "eval_linker_tanimoto_similarity": 0.0,
+      "eval_linker_valid": 0.943791427308882,
+      "eval_loss": 0.6559651494026184,
+      "eval_num_fragments": 3.012372956252762,
+      "eval_poi_equal": 0.0026513477684489617,
+      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
+      "eval_poi_graph_edit_distance_norm": 1.0,
+      "eval_poi_has_attachment_point(s)": 0.2973928413610252,
+      "eval_poi_heavy_atoms_difference": 25.526115775519223,
+      "eval_poi_heavy_atoms_difference_norm": 0.7772625095843685,
+      "eval_poi_tanimoto_similarity": 0.0,
+      "eval_poi_valid": 0.2973928413610252,
+      "eval_reassembly": 0.0,
+      "eval_reassembly_nostereo": 0.0,
+      "eval_runtime": 1361.2908,
+      "eval_samples_per_second": 8.312,
+      "eval_steps_per_second": 0.13,
+      "eval_tanimoto_similarity": 0.0,
+      "eval_valid": 0.2877596111356606,
+      "step": 7500
+    },
+    {
+      "epoch": 0.789110278161373,
+      "grad_norm": 1.2238191366195679,
+      "learning_rate": 1.1633462903366549e-06,
+      "loss": 0.5506,
+      "step": 8000
+    },
+    {
+      "epoch": 0.8384296705464589,
+      "grad_norm": 1.3292711973190308,
+      "learning_rate": 7.667233943405157e-07,
+      "loss": 0.5139,
+      "step": 8500
+    },
+    {
+      "epoch": 0.8877490629315447,
+      "grad_norm": 1.052363634109497,
+      "learning_rate": 3.804131208984208e-07,
+      "loss": 0.4937,
+      "step": 9000
+    },
+    {
+      "epoch": 0.9370684553166305,
+      "grad_norm": 1.1418038606643677,
+      "learning_rate": 1.014774300592085e-07,
+      "loss": 0.4831,
+      "step": 9500
+    },
+    {
+      "epoch": 0.9863878477017163,
+      "grad_norm": 0.9714767932891846,
+      "learning_rate": 0.0,
+      "loss": 0.4799,
+      "step": 10000
+    },
+    {
+      "epoch": 0.9863878477017163,
+      "eval_all_ligands_equal": 0.0,
+      "eval_e3_equal": 0.08590366769774635,
+      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
+      "eval_e3_graph_edit_distance_norm": 1.0,
+      "eval_e3_has_attachment_point(s)": 0.9799381352187362,
+      "eval_e3_heavy_atoms_difference": -0.6433937251436147,
+      "eval_e3_heavy_atoms_difference_norm": -0.060812794107813935,
+      "eval_e3_tanimoto_similarity": 0.0,
+      "eval_e3_valid": 0.9799381352187362,
+      "eval_has_all_attachment_points": 0.8074237737516571,
+      "eval_has_three_substructures": 0.981352187361909,
+      "eval_heavy_atoms_difference": 55.751480335837385,
+      "eval_heavy_atoms_difference_norm": 0.7168565547640369,
+      "eval_linker_equal": 0.0015024304021210783,
+      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
+      "eval_linker_graph_edit_distance_norm": 1.0,
+      "eval_linker_has_attachment_point(s)": 0.9653557224922669,
+      "eval_linker_heavy_atoms_difference": 7.463190455148034,
+      "eval_linker_heavy_atoms_difference_norm": 0.34341321095264715,
+      "eval_linker_tanimoto_similarity": 0.0,
+      "eval_linker_valid": 0.9653557224922669,
+      "eval_loss": 0.5779388546943665,
+      "eval_num_fragments": 3.00415377817057,
+      "eval_poi_equal": 0.011312417145382236,
+      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
+      "eval_poi_graph_edit_distance_norm": 1.0,
+      "eval_poi_has_attachment_point(s)": 0.371011931064958,
+      "eval_poi_heavy_atoms_difference": 23.019619973486524,
+      "eval_poi_heavy_atoms_difference_norm": 0.6868493188139538,
+      "eval_poi_tanimoto_similarity": 0.0,
+      "eval_poi_valid": 0.371011931064958,
+      "eval_reassembly": 0.0,
+      "eval_reassembly_nostereo": 0.0,
+      "eval_runtime": 1413.0201,
+      "eval_samples_per_second": 8.008,
+      "eval_steps_per_second": 0.125,
+      "eval_tanimoto_similarity": 0.0,
+      "eval_valid": 0.3641184268669907,
+      "step": 10000
     }
   ],
   "logging_steps": 500,
@@ -133,12 +287,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 4.33139477262336e+16,
+  "total_flos": 8.663036348367667e+16,
   "train_batch_size": 128,
   "trial_name": "trial-number=21-learning_rate=1.6e-06-warmup_ratio=0.080-num_cycles=1.500",
   "trial_params": {