Training in progress, step 5000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e1d14cd9a14a39a8b5b28b7db1dadcc03089c085e930fb40dc21668ec4a5251f
+oid sha256:09c4e932c0bf0f1f87256c57c99dacd3bbc5e0c2b19c9d5efa347c116a998cbf
 size 409608164

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8bc6e66d3e668e6510ff804c2e4355b2415fc57786584125f8948bb21937440b
+oid sha256:0855748c992337b0aeba1bc7f5ceb80da5ccf7b010550eec8229a041b92244d6
 size 814647162

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3de1adb425137c7cbf77fdc41e85acfd50daeef753f87a52737f8554b5025b18
+oid sha256:f6732c671f36151cdd05b97be18c12b2988cffb1f309a2cb00d544de58eb0ce2
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c54f106895f206c190fccfe0b8538ad9cc159b19305e79ed829d9a63b1b69a5
+oid sha256:472c9aa780093fe594de4986d4ba797952097e39c06067e3fad4b835e8326d81
 size 1192

last-checkpoint/trainer_state.json

@@ -1,278 +1,124 @@
 {
-  "best_metric": 8.837825894829872e-05,
-  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25-rand-smiles/trial-number=5-learning_rate=1.1e-06-warmup_ratio=0.070-min_lr=0.000-factor=0.910/checkpoint-10000",
-  "epoch": 0.9863878477017163,
+  "best_metric": 0.0,
+  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25-rand-smiles/trial-number=6-learning_rate=4.9e-04-warmup_ratio=0.040-min_lr=0.000-factor=0.460/checkpoint-5000",
+  "epoch": 0.49319392385085814,
   "eval_steps": 2500,
-  "global_step": 10000,
+  "global_step": 5000,
   "is_hyper_param_search": true,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.049319392385085814,
-      "grad_norm": 3.6603024005889893,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 4.2002,
+      "grad_norm": 0.6788614988327026,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5969,
       "step": 500
     },
     {
       "epoch": 0.09863878477017163,
-      "grad_norm": 1.9637521505355835,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 1.9187,
+      "grad_norm": 0.7310829758644104,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5675,
       "step": 1000
     },
     {
       "epoch": 0.14795817715525744,
-      "grad_norm": 1.449631929397583,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 1.4159,
+      "grad_norm": 0.7336956858634949,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5712,
       "step": 1500
     },
     {
       "epoch": 0.19727756954034326,
-      "grad_norm": 1.434281826019287,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 1.2066,
+      "grad_norm": 0.9253264665603638,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5666,
       "step": 2000
     },
     {
       "epoch": 0.24659696192542907,
-      "grad_norm": 1.311953067779541,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 1.0605,
+      "grad_norm": 0.8935646414756775,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5663,
       "step": 2500
     },
     {
       "epoch": 0.2959163543105149,
-      "grad_norm": 1.3984962701797485,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.9508,
+      "grad_norm": 0.749767541885376,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5635,
       "step": 3000
     },
     {
       "epoch": 0.3452357466956007,
-      "grad_norm": 1.29249906539917,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.8594,
+      "grad_norm": 0.9100434184074402,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5616,
       "step": 3500
     },
     {
       "epoch": 0.3945551390806865,
-      "grad_norm": 1.4493184089660645,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.786,
+      "grad_norm": 0.9583754539489746,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5619,
       "step": 4000
     },
     {
       "epoch": 0.4438745314657723,
-      "grad_norm": 1.42921781539917,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.7204,
+      "grad_norm": 1.0354633331298828,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5594,
       "step": 4500
     },
     {
       "epoch": 0.49319392385085814,
-      "grad_norm": 1.4308568239212036,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.6681,
+      "grad_norm": 1.0010348558425903,
+      "learning_rate": 0.0004854655346216474,
+      "loss": 3.5595,
       "step": 5000
     },
     {
       "epoch": 0.49319392385085814,
       "eval_all_ligands_equal": 0.0,
-      "eval_e3_equal": 0.04348210340256297,
+      "eval_e3_equal": 0.0,
       "eval_e3_graph_edit_distance": 9.999999999999999e+63,
       "eval_e3_graph_edit_distance_norm": 1.0,
-      "eval_e3_has_attachment_point(s)": 0.9496243923994697,
-      "eval_e3_heavy_atoms_difference": 1.6608926204153778,
-      "eval_e3_heavy_atoms_difference_norm": 0.007137613196833208,
+      "eval_e3_has_attachment_point(s)": 0.0,
+      "eval_e3_heavy_atoms_difference": 28.086433937251435,
+      "eval_e3_heavy_atoms_difference_norm": 1.0,
       "eval_e3_tanimoto_similarity": 0.0,
-      "eval_e3_valid": 0.9496243923994697,
-      "eval_has_all_attachment_points": 0.6837825894829872,
-      "eval_has_three_substructures": 0.950596553247901,
-      "eval_heavy_atoms_difference": 64.6774193548387,
-      "eval_heavy_atoms_difference_norm": 0.8393433147141454,
-      "eval_linker_equal": 0.0013256738842244808,
+      "eval_e3_valid": 0.0,
+      "eval_has_all_attachment_points": 0.0,
+      "eval_has_three_substructures": 0.0,
+      "eval_heavy_atoms_difference": 75.67370746796288,
+      "eval_heavy_atoms_difference_norm": 1.0,
+      "eval_linker_equal": 0.0,
       "eval_linker_graph_edit_distance": 9.999999999999999e+63,
       "eval_linker_graph_edit_distance_norm": 1.0,
-      "eval_linker_has_attachment_point(s)": 0.9260273972602739,
-      "eval_linker_heavy_atoms_difference": 9.002651347768449,
-      "eval_linker_heavy_atoms_difference_norm": 0.45586405473119984,
+      "eval_linker_has_attachment_point(s)": 0.0,
+      "eval_linker_heavy_atoms_difference": 16.430402121078213,
+      "eval_linker_heavy_atoms_difference_norm": 1.0,
       "eval_linker_tanimoto_similarity": 0.0,
-      "eval_linker_valid": 0.9260273972602739,
-      "eval_loss": 0.7126381993293762,
-      "eval_num_fragments": 3.0372956252761822,
-      "eval_poi_equal": 0.00017675651789659743,
+      "eval_linker_valid": 0.0,
+      "eval_loss": 3.582303047180176,
+      "eval_num_fragments": 1.0024745912505524,
+      "eval_poi_equal": 0.0,
       "eval_poi_graph_edit_distance": 9.999999999999999e+63,
       "eval_poi_graph_edit_distance_norm": 1.0,
-      "eval_poi_has_attachment_point(s)": 0.29854175872735306,
-      "eval_poi_heavy_atoms_difference": 25.908882015024304,
-      "eval_poi_heavy_atoms_difference_norm": 0.7922565446145508,
+      "eval_poi_has_attachment_point(s)": 0.0,
+      "eval_poi_heavy_atoms_difference": 31.15687140963323,
+      "eval_poi_heavy_atoms_difference_norm": 1.0,
       "eval_poi_tanimoto_similarity": 0.0,
-      "eval_poi_valid": 0.29854175872735306,
+      "eval_poi_valid": 0.0,
       "eval_reassembly": 0.0,
       "eval_reassembly_nostereo": 0.0,
-      "eval_runtime": 1368.1465,
-      "eval_samples_per_second": 8.27,
-      "eval_steps_per_second": 0.129,
+      "eval_runtime": 5688.6938,
+      "eval_samples_per_second": 1.989,
+      "eval_steps_per_second": 0.031,
       "eval_tanimoto_similarity": 0.0,
-      "eval_valid": 0.2896155545735749,
+      "eval_valid": 0.0,
       "step": 5000
-    },
-    {
-      "epoch": 0.542513316235944,
-      "grad_norm": 1.2828806638717651,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.6206,
-      "step": 5500
-    },
-    {
-      "epoch": 0.5918327086210298,
-      "grad_norm": 1.2522315979003906,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.5797,
-      "step": 6000
-    },
-    {
-      "epoch": 0.6411521010061156,
-      "grad_norm": 1.1082614660263062,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.5435,
-      "step": 6500
-    },
-    {
-      "epoch": 0.6904714933912014,
-      "grad_norm": 1.3583234548568726,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.5116,
-      "step": 7000
-    },
-    {
-      "epoch": 0.7397908857762873,
-      "grad_norm": 1.1005616188049316,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.4817,
-      "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)": 0.984003535130358,
-      "eval_e3_heavy_atoms_difference": -0.5573133009279717,
-      "eval_e3_heavy_atoms_difference_norm": -0.06045991079256801,
-      "eval_e3_tanimoto_similarity": 0.0,
-      "eval_e3_valid": 0.984003535130358,
-      "eval_has_all_attachment_points": 0.8299602297834733,
-      "eval_has_three_substructures": 0.9849756959787892,
-      "eval_heavy_atoms_difference": 52.16049491825011,
-      "eval_heavy_atoms_difference_norm": 0.6685409446631507,
-      "eval_linker_equal": 0.0017675651789659744,
-      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
-      "eval_linker_graph_edit_distance_norm": 1.0,
-      "eval_linker_has_attachment_point(s)": 0.972779496243924,
-      "eval_linker_heavy_atoms_difference": 6.889527176314626,
-      "eval_linker_heavy_atoms_difference_norm": 0.30243755302342107,
-      "eval_linker_tanimoto_similarity": 0.0,
-      "eval_linker_valid": 0.972779496243924,
-      "eval_loss": 0.5654022097587585,
-      "eval_num_fragments": 3.004418912947415,
-      "eval_poi_equal": 0.011224038886433937,
-      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
-      "eval_poi_graph_edit_distance_norm": 1.0,
-      "eval_poi_has_attachment_point(s)": 0.420415377817057,
-      "eval_poi_heavy_atoms_difference": 21.88634555899249,
-      "eval_poi_heavy_atoms_difference_norm": 0.6511884108724427,
-      "eval_poi_tanimoto_similarity": 0.0,
-      "eval_poi_valid": 0.420415377817057,
-      "eval_reassembly": 0.0,
-      "eval_reassembly_nostereo": 0.0,
-      "eval_runtime": 1418.3103,
-      "eval_samples_per_second": 7.978,
-      "eval_steps_per_second": 0.125,
-      "eval_tanimoto_similarity": 0.0,
-      "eval_valid": 0.4154661953159523,
-      "step": 7500
-    },
-    {
-      "epoch": 0.789110278161373,
-      "grad_norm": 1.2720683813095093,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.4566,
-      "step": 8000
-    },
-    {
-      "epoch": 0.8384296705464589,
-      "grad_norm": 1.2321172952651978,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.4311,
-      "step": 8500
-    },
-    {
-      "epoch": 0.8877490629315447,
-      "grad_norm": 1.034532070159912,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.4106,
-      "step": 9000
-    },
-    {
-      "epoch": 0.9370684553166305,
-      "grad_norm": 1.1301201581954956,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.3902,
-      "step": 9500
-    },
-    {
-      "epoch": 0.9863878477017163,
-      "grad_norm": 0.7160457372665405,
-      "learning_rate": 1.1386440825931043e-06,
-      "loss": 0.3717,
-      "step": 10000
-    },
-    {
-      "epoch": 0.9863878477017163,
-      "eval_all_ligands_equal": 8.837825894829872e-05,
-      "eval_e3_equal": 0.11391957578435705,
-      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
-      "eval_e3_graph_edit_distance_norm": 1.0,
-      "eval_e3_has_attachment_point(s)": 0.9894829871851525,
-      "eval_e3_heavy_atoms_difference": -1.1898365002209457,
-      "eval_e3_heavy_atoms_difference_norm": -0.07529202188483206,
-      "eval_e3_tanimoto_similarity": 0.0,
-      "eval_e3_valid": 0.9894829871851525,
-      "eval_has_all_attachment_points": 0.8728236853733982,
-      "eval_has_three_substructures": 0.9906319045514803,
-      "eval_heavy_atoms_difference": 40.78983650022094,
-      "eval_heavy_atoms_difference_norm": 0.5206536331774743,
-      "eval_linker_equal": 0.0026513477684489617,
-      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
-      "eval_linker_graph_edit_distance_norm": 1.0,
-      "eval_linker_has_attachment_point(s)": 0.984003535130358,
-      "eval_linker_heavy_atoms_difference": 6.854794520547945,
-      "eval_linker_heavy_atoms_difference_norm": 0.30805691525777035,
-      "eval_linker_tanimoto_similarity": 0.0,
-      "eval_linker_valid": 0.984003535130358,
-      "eval_loss": 0.48492810130119324,
-      "eval_num_fragments": 2.9957578435704817,
-      "eval_poi_equal": 0.03729562527618206,
-      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
-      "eval_poi_graph_edit_distance_norm": 1.0,
-      "eval_poi_has_attachment_point(s)": 0.5657976137870084,
-      "eval_poi_heavy_atoms_difference": 16.954043305346886,
-      "eval_poi_heavy_atoms_difference_norm": 0.4893511561351334,
-      "eval_poi_tanimoto_similarity": 0.0,
-      "eval_poi_valid": 0.5657976137870084,
-      "eval_reassembly": 8.837825894829872e-05,
-      "eval_reassembly_nostereo": 8.837825894829872e-05,
-      "eval_runtime": 1451.4515,
-      "eval_samples_per_second": 7.796,
-      "eval_steps_per_second": 0.122,
-      "eval_tanimoto_similarity": 0.0,
-      "eval_valid": 0.5629695095006628,
-      "step": 10000
     }
   ],
   "logging_steps": 500,
@@ -287,18 +133,18 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 8.729673191023411e+16,
+  "total_flos": 4.364959997072179e+16,
   "train_batch_size": 128,
-  "trial_name": "trial-number=5-learning_rate=1.1e-06-warmup_ratio=0.070-min_lr=0.000-factor=0.910",
+  "trial_name": "trial-number=6-learning_rate=4.9e-04-warmup_ratio=0.040-min_lr=0.000-factor=0.460",
   "trial_params": {
-    "factor": 0.91,
-    "learning_rate": 1.1386440825931043e-06,
-    "min_lr": 1.1386082331680292e-12,
-    "warmup_ratio": 0.06999999999999999
+    "factor": 0.45999999999999996,
+    "learning_rate": 0.0004854655346216474,
+    "min_lr": 5.158242893359211e-10,
+    "warmup_ratio": 0.04
   }
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a378ac23f30a31ba156b2d137739247698d58210a372732645c0744c38de224f
+oid sha256:b95041f6a010f2cf869dc8bc44f5902cbc00a43c6932510edd799e444f572a41
 size 7544