Training in progress, step 10000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c3d131116cfd4dd25bc1b0e3efe04b5e5a01f59ac4c2711f23f79391aedbd33
+oid sha256:f6b9fcea6d6d61860f983703a2d65e182d32b7cf4bd5a62beb9ec8eb364ffe8e
 size 409608164

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8782898d0392c681104668893c5658959694b1d08fa04592ce90312a858cce6
+oid sha256:00331de6fb70ff1e930f9d5f1cf17e3f5c1dc1888dc71f7e6607c345cbbdea9f
 size 814647162

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b4bdb01f4cdc76d9e5349316151e3fe8f0abf2240276e995d1a75b29c52b47c
+oid sha256:3963bd91625aaeb7bd19504b27420f7a90a321a65cce4576e5cb8f91af1b0a9e
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f693bf51c2c055b152f890bd5d50e563d8d32ff1954a25492f6a9c28ce352e2
+oid sha256:04c72869181363c7c0dcffb7a71857fd6a7c81a93e70aff2985fc26363295969
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.4823831444759207,
-  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_cosine-opt25-rand-smiles/checkpoint-5000",
-  "epoch": 0.49319392385085814,
+  "best_metric": 0.5389518413597734,
+  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_cosine-opt25-rand-smiles/checkpoint-10000",
+  "epoch": 0.9863878477017163,
   "eval_steps": 2500,
-  "global_step": 5000,
+  "global_step": 10000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -119,6 +119,160 @@
       "eval_tanimoto_similarity": 0.0,
       "eval_valid": 0.9440509915014165,
       "step": 5000
+    },
+    {
+      "epoch": 0.542513316235944,
+      "grad_norm": 0.23216110467910767,
+      "learning_rate": 2.372969012074575e-05,
+      "loss": 0.0101,
+      "step": 5500
+    },
+    {
+      "epoch": 0.5918327086210298,
+      "grad_norm": 0.14501763880252838,
+      "learning_rate": 1.9551116361503457e-05,
+      "loss": 0.0089,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6411521010061156,
+      "grad_norm": 0.12931592762470245,
+      "learning_rate": 1.55275867959039e-05,
+      "loss": 0.0078,
+      "step": 6500
+    },
+    {
+      "epoch": 0.6904714933912014,
+      "grad_norm": 0.1443043202161789,
+      "learning_rate": 1.1773588158934163e-05,
+      "loss": 0.0072,
+      "step": 7000
+    },
+    {
+      "epoch": 0.7397908857762873,
+      "grad_norm": 0.13492755591869354,
+      "learning_rate": 8.395937870495241e-06,
+      "loss": 0.0064,
+      "step": 7500
+    },
+    {
+      "epoch": 0.7397908857762873,
+      "eval_all_ligands_equal": 0.5282400849858357,
+      "eval_e3_equal": 0.8016997167138811,
+      "eval_e3_graph_edit_distance": Infinity,
+      "eval_e3_graph_edit_distance_norm": Infinity,
+      "eval_e3_has_attachment_point(s)": 0.9944228045325779,
+      "eval_e3_heavy_atoms_difference": 0.342864730878187,
+      "eval_e3_heavy_atoms_difference_norm": 0.0053395654721611,
+      "eval_e3_tanimoto_similarity": 0.0,
+      "eval_e3_valid": 0.9944228045325779,
+      "eval_has_all_attachment_points": 0.9873406515580736,
+      "eval_has_three_substructures": 0.9984950424929179,
+      "eval_heavy_atoms_difference": 4.519475920679887,
+      "eval_heavy_atoms_difference_norm": 0.059262409138829596,
+      "eval_linker_equal": 0.7532754957507082,
+      "eval_linker_graph_edit_distance": 4.0722379603399434e+61,
+      "eval_linker_graph_edit_distance_norm": 0.04522498623517566,
+      "eval_linker_has_attachment_point(s)": 0.99592776203966,
+      "eval_linker_heavy_atoms_difference": 0.2563739376770538,
+      "eval_linker_heavy_atoms_difference_norm": -2.297886744053873e-06,
+      "eval_linker_tanimoto_similarity": 0.0,
+      "eval_linker_valid": 0.99592776203966,
+      "eval_loss": 0.3155761659145355,
+      "eval_num_fragments": 3.000265580736544,
+      "eval_poi_equal": 0.7551345609065155,
+      "eval_poi_graph_edit_distance": Infinity,
+      "eval_poi_graph_edit_distance_norm": Infinity,
+      "eval_poi_has_attachment_point(s)": 0.9531692634560907,
+      "eval_poi_heavy_atoms_difference": 1.3395007082152974,
+      "eval_poi_heavy_atoms_difference_norm": 0.03956240679400847,
+      "eval_poi_tanimoto_similarity": 0.0,
+      "eval_poi_valid": 0.9531692634560907,
+      "eval_reassembly": 0.5359419263456091,
+      "eval_reassembly_nostereo": 0.5688739376770539,
+      "eval_runtime": 2393.2072,
+      "eval_samples_per_second": 4.72,
+      "eval_steps_per_second": 0.074,
+      "eval_tanimoto_similarity": 0.0,
+      "eval_valid": 0.9476805949008499,
+      "step": 7500
+    },
+    {
+      "epoch": 0.789110278161373,
+      "grad_norm": 0.12762106955051422,
+      "learning_rate": 5.490744620021837e-06,
+      "loss": 0.0059,
+      "step": 8000
+    },
+    {
+      "epoch": 0.8384296705464589,
+      "grad_norm": 0.12324267625808716,
+      "learning_rate": 3.1406736605785858e-06,
+      "loss": 0.0056,
+      "step": 8500
+    },
+    {
+      "epoch": 0.8877490629315447,
+      "grad_norm": 0.10565518587827682,
+      "learning_rate": 1.412594626586694e-06,
+      "loss": 0.0053,
+      "step": 9000
+    },
+    {
+      "epoch": 0.9370684553166305,
+      "grad_norm": 0.10529914498329163,
+      "learning_rate": 3.5567880489165705e-07,
+      "loss": 0.0052,
+      "step": 9500
+    },
+    {
+      "epoch": 0.9863878477017163,
+      "grad_norm": 0.11241015791893005,
+      "learning_rate": 0.0,
+      "loss": 0.0051,
+      "step": 10000
+    },
+    {
+      "epoch": 0.9863878477017163,
+      "eval_all_ligands_equal": 0.5389518413597734,
+      "eval_e3_equal": 0.8036473087818697,
+      "eval_e3_graph_edit_distance": Infinity,
+      "eval_e3_graph_edit_distance_norm": Infinity,
+      "eval_e3_has_attachment_point(s)": 0.9966359773371105,
+      "eval_e3_heavy_atoms_difference": 0.29762747875354106,
+      "eval_e3_heavy_atoms_difference_norm": 0.0033101986712786185,
+      "eval_e3_tanimoto_similarity": 0.0,
+      "eval_e3_valid": 0.9966359773371105,
+      "eval_has_all_attachment_points": 0.9905276203966006,
+      "eval_has_three_substructures": 0.9992032577903682,
+      "eval_heavy_atoms_difference": 4.104904390934844,
+      "eval_heavy_atoms_difference_norm": 0.0535843946056828,
+      "eval_linker_equal": 0.7668201133144475,
+      "eval_linker_graph_edit_distance": 2.8328611898017e+61,
+      "eval_linker_graph_edit_distance_norm": Infinity,
+      "eval_linker_has_attachment_point(s)": 0.9971671388101983,
+      "eval_linker_heavy_atoms_difference": 0.3031161473087819,
+      "eval_linker_heavy_atoms_difference_norm": 0.0033820491296884894,
+      "eval_linker_tanimoto_similarity": 0.0,
+      "eval_linker_valid": 0.9971671388101983,
+      "eval_loss": 0.31840986013412476,
+      "eval_num_fragments": 3.000265580736544,
+      "eval_poi_equal": 0.7619511331444759,
+      "eval_poi_graph_edit_distance": Infinity,
+      "eval_poi_graph_edit_distance_norm": Infinity,
+      "eval_poi_has_attachment_point(s)": 0.9589235127478754,
+      "eval_poi_heavy_atoms_difference": 1.1856409348441925,
+      "eval_poi_heavy_atoms_difference_norm": 0.035200135322798254,
+      "eval_poi_tanimoto_similarity": 0.0,
+      "eval_poi_valid": 0.9589235127478754,
+      "eval_reassembly": 0.5459454674220963,
+      "eval_reassembly_nostereo": 0.5798512747875354,
+      "eval_runtime": 2383.5536,
+      "eval_samples_per_second": 4.739,
+      "eval_steps_per_second": 0.074,
+      "eval_tanimoto_similarity": 0.0,
+      "eval_valid": 0.9546742209631728,
+      "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": 4.364959997072179e+16,
+  "total_flos": 8.729673191023411e+16,
   "train_batch_size": 128,
   "trial_name": null,
   "trial_params": null