Training in progress, step 10000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa62b5efb1844ae53dd7f45956a6677f10bce73ad71356720907215536556f68
+oid sha256:ea6b2a63e760eff0d45b2d57870a9cdfe6e9e18695c39d9e7e147b4c3d883ea4
 size 409608164

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc04c3f034f1b0d3fdd074893ec0e5aa7b2b5e5044a133b359c4677f487fa79c
+oid sha256:4d006bcf90d1175c35fa2c258bacd4ca60d14976391a619e9ba60990d64dd91e
 size 814647162

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7944a123a20e716aab97dbd8cc4190b23fa0da33c6b81e437bd160e8a6452078
+oid sha256:e986d46d706b3fb91b88a850b06182e639890487ae4440c3df7feb76772997f1
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e485a1d5071b23b811f4a1b96aeaa0547c3f98de0681f55bcfba2da94aaa8a49
+oid sha256:f33d61e7faf50026c6df62b1820a4e5e991131d1b2a9f07f2595d907464c4ed2
 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-rand-smiles/trial-number=12-learning_rate=2.2e-06-warmup_ratio=0.050-num_cycles=10.000/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": 0.2844896155545736,
       "step": 5000
+    },
+    {
+      "epoch": 0.542513316235944,
+      "grad_norm": 1.2575222253799438,
+      "learning_rate": 1.6656876679399406e-06,
+      "loss": 0.6502,
+      "step": 5500
+    },
+    {
+      "epoch": 0.5918327086210298,
+      "grad_norm": 1.1440157890319824,
+      "learning_rate": 7.501476710478152e-07,
+      "loss": 0.5992,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6411521010061156,
+      "grad_norm": 1.0409629344940186,
+      "learning_rate": 1.1807005624117962e-06,
+      "loss": 0.5677,
+      "step": 6500
+    },
+    {
+      "epoch": 0.6904714933912014,
+      "grad_norm": 1.311508059501648,
+      "learning_rate": 1.2590240506504755e-06,
+      "loss": 0.5325,
+      "step": 7000
+    },
+    {
+      "epoch": 0.7397908857762873,
+      "grad_norm": 0.9935227632522583,
+      "learning_rate": 6.761699700853001e-07,
+      "loss": 0.5017,
+      "step": 7500
+    },
+    {
+      "epoch": 0.7397908857762873,
+      "eval_all_ligands_equal": 0.0,
+      "eval_e3_equal": 0.0827220503756076,
+      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
+      "eval_e3_graph_edit_distance_norm": 1.0,
+      "eval_e3_has_attachment_point(s)": 0.9852408307556341,
+      "eval_e3_heavy_atoms_difference": -0.6676977463543968,
+      "eval_e3_heavy_atoms_difference_norm": -0.06390924999730649,
+      "eval_e3_tanimoto_similarity": 0.0,
+      "eval_e3_valid": 0.9852408307556341,
+      "eval_has_all_attachment_points": 0.8275740167918691,
+      "eval_has_three_substructures": 0.9867432611577552,
+      "eval_heavy_atoms_difference": 53.99496243923995,
+      "eval_heavy_atoms_difference_norm": 0.692721967350773,
+      "eval_linker_equal": 0.0020326999558108703,
+      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
+      "eval_linker_graph_edit_distance_norm": 1.0,
+      "eval_linker_has_attachment_point(s)": 0.9718073353954927,
+      "eval_linker_heavy_atoms_difference": 7.161290322580645,
+      "eval_linker_heavy_atoms_difference_norm": 0.31956488397760113,
+      "eval_linker_tanimoto_similarity": 0.0,
+      "eval_linker_valid": 0.9718073353954927,
+      "eval_loss": 0.5782048106193542,
+      "eval_num_fragments": 3.0007070260715865,
+      "eval_poi_equal": 0.011489173663278833,
+      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
+      "eval_poi_graph_edit_distance_norm": 1.0,
+      "eval_poi_has_attachment_point(s)": 0.3986743261157755,
+      "eval_poi_heavy_atoms_difference": 22.559699513919576,
+      "eval_poi_heavy_atoms_difference_norm": 0.672502934449603,
+      "eval_poi_tanimoto_similarity": 0.0,
+      "eval_poi_valid": 0.3986743261157755,
+      "eval_reassembly": 0.0,
+      "eval_reassembly_nostereo": 0.0,
+      "eval_runtime": 1422.689,
+      "eval_samples_per_second": 7.953,
+      "eval_steps_per_second": 0.124,
+      "eval_tanimoto_similarity": 0.0,
+      "eval_valid": 0.3926646045072912,
+      "step": 7500
+    },
+    {
+      "epoch": 0.789110278161373,
+      "grad_norm": 1.2995916604995728,
+      "learning_rate": 1.7312149208305083e-06,
+      "loss": 0.4783,
+      "step": 8000
+    },
+    {
+      "epoch": 0.8384296705464589,
+      "grad_norm": 1.1670758724212646,
+      "learning_rate": 2.625183839161817e-07,
+      "loss": 0.4483,
+      "step": 8500
+    },
+    {
+      "epoch": 0.8877490629315447,
+      "grad_norm": 1.0538517236709595,
+      "learning_rate": 2.0633757196738724e-06,
+      "loss": 0.4319,
+      "step": 9000
+    },
+    {
+      "epoch": 0.9370684553166305,
+      "grad_norm": 1.168967843055725,
+      "learning_rate": 3.027835050239297e-08,
+      "loss": 0.4054,
+      "step": 9500
+    },
+    {
+      "epoch": 0.9863878477017163,
+      "grad_norm": 0.8473185300827026,
+      "learning_rate": 2.182809124107808e-06,
+      "loss": 0.3918,
+      "step": 10000
+    },
+    {
+      "epoch": 0.9863878477017163,
+      "eval_all_ligands_equal": 0.0,
+      "eval_e3_equal": 0.11321254971277066,
+      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
+      "eval_e3_graph_edit_distance_norm": 1.0,
+      "eval_e3_has_attachment_point(s)": 0.9885992045956694,
+      "eval_e3_heavy_atoms_difference": -1.165090587715422,
+      "eval_e3_heavy_atoms_difference_norm": -0.0760775265425178,
+      "eval_e3_tanimoto_similarity": 0.0,
+      "eval_e3_valid": 0.9885992045956694,
+      "eval_has_all_attachment_points": 0.8649580203269995,
+      "eval_has_three_substructures": 0.9895713654441007,
+      "eval_heavy_atoms_difference": 42.39593460008838,
+      "eval_heavy_atoms_difference_norm": 0.5396815683959801,
+      "eval_linker_equal": 0.0021210782147591694,
+      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
+      "eval_linker_graph_edit_distance_norm": 1.0,
+      "eval_linker_has_attachment_point(s)": 0.9826778612461334,
+      "eval_linker_heavy_atoms_difference": 6.7108263367211665,
+      "eval_linker_heavy_atoms_difference_norm": 0.2914348823794422,
+      "eval_linker_tanimoto_similarity": 0.0,
+      "eval_linker_valid": 0.9826778612461334,
+      "eval_loss": 0.5012893080711365,
+      "eval_num_fragments": 2.996818382677861,
+      "eval_poi_equal": 0.03049049933716306,
+      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
+      "eval_poi_graph_edit_distance_norm": 1.0,
+      "eval_poi_has_attachment_point(s)": 0.5458241272646929,
+      "eval_poi_heavy_atoms_difference": 18.0181175430844,
+      "eval_poi_heavy_atoms_difference_norm": 0.519674391987532,
+      "eval_poi_tanimoto_similarity": 0.0,
+      "eval_poi_valid": 0.5458241272646929,
+      "eval_reassembly": 0.0,
+      "eval_reassembly_nostereo": 0.0,
+      "eval_runtime": 1411.5279,
+      "eval_samples_per_second": 8.016,
+      "eval_steps_per_second": 0.125,
+      "eval_tanimoto_similarity": 0.0,
+      "eval_valid": 0.5427308882015024,
+      "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": "trial-number=12-learning_rate=2.2e-06-warmup_ratio=0.050-num_cycles=10.000",
   "trial_params": {