Training in progress, step 5000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e7fe164acee2bf98394cc1f6ca77bde3bf4fe3c18a172150d217aa0744f7c0d
+oid sha256:d4f84b63c76ad39c4d29b5e1779cdb4ddf62f0fcbbfc8c6195f3d5135f1c7187
 size 409608164

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:982416f84073374fbfc9bc102c3b314f16c390ffafe0595ec3c0a1bc9bb2bdbc
+oid sha256:bf2ca9e924d1cab355d6dcbab707f5dbf8b39eed6e13d75f60b419650e16b97c
 size 814647162

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59e8ca36bfab9e84d90a7d949d509b7af9d968ff1c66c18555967285cddf14a3
+oid sha256:b925556f82b124035f3ca7135a752041c7e1c8d324e6a65157a89e3b81097bc3
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d150c531f1bb84bb58605a639401cc62d9ee0f1aeab46fc43a18748a772f2c54
+oid sha256:4adbc58e93a831eb642bd288aafadcc9d5456925c59a14e4d584eec75bb8c5c3
 size 1192

last-checkpoint/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.39628811312417145,
-  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25/trial-number=0-learning_rate=2.0e-05-warmup_ratio=0.060-min_lr=0.000-factor=0.910/checkpoint-5000",
+  "best_metric": 0.4479010163499779,
+  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25/trial-number=1-learning_rate=2.5e-05-warmup_ratio=0.060-min_lr=0.000-factor=0.810/checkpoint-5000",
   "epoch": 0.49319392385085814,
   "eval_steps": 2500,
   "global_step": 5000,
@@ -10,114 +10,114 @@
   "log_history": [
     {
       "epoch": 0.049319392385085814,
-      "grad_norm": 0.9932323098182678,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 1.3472,
+      "grad_norm": 0.8631236553192139,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 1.2428,
       "step": 500
     },
     {
       "epoch": 0.09863878477017163,
-      "grad_norm": 0.6675044894218445,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 0.3791,
+      "grad_norm": 0.6576346158981323,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 0.3154,
       "step": 1000
     },
     {
       "epoch": 0.14795817715525744,
-      "grad_norm": 0.5713032484054565,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 0.197,
+      "grad_norm": 0.5040571093559265,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 0.1531,
       "step": 1500
     },
     {
       "epoch": 0.19727756954034326,
-      "grad_norm": 0.4924110472202301,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 0.1224,
+      "grad_norm": 0.47591033577919006,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 0.0904,
       "step": 2000
     },
     {
       "epoch": 0.24659696192542907,
-      "grad_norm": 0.38072213530540466,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 0.0814,
+      "grad_norm": 0.3052867352962494,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 0.0588,
       "step": 2500
     },
     {
       "epoch": 0.2959163543105149,
-      "grad_norm": 0.2933690547943115,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 0.0578,
+      "grad_norm": 0.26361268758773804,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 0.0418,
       "step": 3000
     },
     {
       "epoch": 0.3452357466956007,
-      "grad_norm": 0.24898405373096466,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 0.0432,
+      "grad_norm": 0.23498503863811493,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 0.0308,
       "step": 3500
     },
     {
       "epoch": 0.3945551390806865,
-      "grad_norm": 0.27284932136535645,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 0.0338,
+      "grad_norm": 0.2148304432630539,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 0.0245,
       "step": 4000
     },
     {
       "epoch": 0.4438745314657723,
-      "grad_norm": 0.24178475141525269,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 0.027,
+      "grad_norm": 0.19326996803283691,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 0.0193,
       "step": 4500
     },
     {
       "epoch": 0.49319392385085814,
-      "grad_norm": 0.2615422010421753,
-      "learning_rate": 1.9906996673933362e-05,
-      "loss": 0.0224,
+      "grad_norm": 0.23251751065254211,
+      "learning_rate": 2.4519131356633107e-05,
+      "loss": 0.0162,
       "step": 5000
     },
     {
       "epoch": 0.49319392385085814,
-      "eval_all_ligands_equal": 0.39628811312417145,
-      "eval_e3_equal": 0.7779054352629253,
+      "eval_all_ligands_equal": 0.4479010163499779,
+      "eval_e3_equal": 0.7840035351303579,
       "eval_e3_graph_edit_distance": 9.999999999999999e+63,
       "eval_e3_graph_edit_distance_norm": 1.0,
-      "eval_e3_has_attachment_point(s)": 0.9931064958020327,
-      "eval_e3_heavy_atoms_difference": 0.30676093680954486,
-      "eval_e3_heavy_atoms_difference_norm": 0.002422775502318104,
+      "eval_e3_has_attachment_point(s)": 0.9891294741493593,
+      "eval_e3_heavy_atoms_difference": 0.4326999558108705,
+      "eval_e3_heavy_atoms_difference_norm": 0.00617409936456833,
       "eval_e3_tanimoto_similarity": 0.0,
-      "eval_e3_valid": 0.9931064958020327,
-      "eval_has_all_attachment_points": 0.9843570481661511,
-      "eval_has_three_substructures": 0.9992929739284137,
-      "eval_heavy_atoms_difference": 8.062218294299603,
-      "eval_heavy_atoms_difference_norm": 0.10429584967389881,
-      "eval_linker_equal": 0.5711886875828546,
+      "eval_e3_valid": 0.9891294741493593,
+      "eval_has_all_attachment_points": 0.9788775961113566,
+      "eval_has_three_substructures": 0.9994697304463102,
+      "eval_heavy_atoms_difference": 8.114007954043306,
+      "eval_heavy_atoms_difference_norm": 0.10592343628762439,
+      "eval_linker_equal": 0.636411842686699,
       "eval_linker_graph_edit_distance": 9.999999999999999e+63,
       "eval_linker_graph_edit_distance_norm": 1.0,
-      "eval_linker_has_attachment_point(s)": 0.9961997348652232,
-      "eval_linker_heavy_atoms_difference": 0.9341581970835174,
-      "eval_linker_heavy_atoms_difference_norm": 0.024806250123980716,
+      "eval_linker_has_attachment_point(s)": 0.9965532479010164,
+      "eval_linker_heavy_atoms_difference": 0.6443658859920459,
+      "eval_linker_heavy_atoms_difference_norm": 0.010009960469603861,
       "eval_linker_tanimoto_similarity": 0.0,
-      "eval_linker_valid": 0.9961997348652232,
-      "eval_loss": 0.28600096702575684,
-      "eval_num_fragments": 2.999646486964207,
-      "eval_poi_equal": 0.70234202386213,
+      "eval_linker_valid": 0.9965532479010164,
+      "eval_loss": 0.2841632664203644,
+      "eval_num_fragments": 3.0001767565178965,
+      "eval_poi_equal": 0.7266460450729121,
       "eval_poi_graph_edit_distance": 9.999999999999999e+63,
       "eval_poi_graph_edit_distance_norm": 1.0,
-      "eval_poi_has_attachment_point(s)": 0.9204595669465312,
-      "eval_poi_heavy_atoms_difference": 2.6359699513919574,
-      "eval_poi_heavy_atoms_difference_norm": 0.07909068566841335,
+      "eval_poi_has_attachment_point(s)": 0.9194874060980999,
+      "eval_poi_heavy_atoms_difference": 2.440300486080424,
+      "eval_poi_heavy_atoms_difference_norm": 0.07726015738826272,
       "eval_poi_tanimoto_similarity": 0.0,
-      "eval_poi_valid": 0.9204595669465312,
-      "eval_reassembly": 0.40247459125055235,
-      "eval_reassembly_nostereo": 0.4217410517012815,
-      "eval_runtime": 1470.3159,
-      "eval_samples_per_second": 7.696,
-      "eval_steps_per_second": 0.12,
+      "eval_poi_valid": 0.9194874060980999,
+      "eval_reassembly": 0.4547945205479452,
+      "eval_reassembly_nostereo": 0.4756517896597437,
+      "eval_runtime": 1574.3525,
+      "eval_samples_per_second": 7.187,
+      "eval_steps_per_second": 0.112,
       "eval_tanimoto_similarity": 0.0,
-      "eval_valid": 0.9122403888643393,
+      "eval_valid": 0.9072912063632347,
       "step": 5000
     }
   ],
@@ -140,11 +140,11 @@
   },
   "total_flos": 4.33139477262336e+16,
   "train_batch_size": 128,
-  "trial_name": "trial-number=0-learning_rate=2.0e-05-warmup_ratio=0.060-min_lr=0.000-factor=0.910",
+  "trial_name": "trial-number=1-learning_rate=2.5e-05-warmup_ratio=0.060-min_lr=0.000-factor=0.810",
   "trial_params": {
-    "factor": 0.91,
-    "learning_rate": 1.9906996673933362e-05,
-    "min_lr": 2.910635913133059e-09,
+    "factor": 0.81,
+    "learning_rate": 2.4519131356633107e-05,
+    "min_lr": 4.1003934714656736e-09,
     "warmup_ratio": 0.06
   }
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed9fa2c5779e79b2d5ae66e6d1c4f9a70b8ce9f13125f8a9d13bcd0dfdaa4af9
+oid sha256:3eec7615192e1bb866379e3b01579890ccadb1f12990ad30b73a3999ff22d577
 size 7480