Upload folder using huggingface_hub

training_checkpoints/checkpoint-2572/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93d9c2cbbb24256937c959c3664e210ae1e2e4fa9b8e7b1fa9e9445d2653dcf2
+oid sha256:d95af8074cc971fb6c23ecc1511bf1b9b21b45aa5df5575a187b84c28670ca6e
 size 498652812

training_checkpoints/checkpoint-2572/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5df5400ab439a281f15f6aebc703d1f4668e94f31de2dc314862f8fa26fbf8fe
+oid sha256:6d592eaee987723e13c4d9be2f1d1433eda3e5b8d1d4479731698997bc3af11a
 size 997425402

training_checkpoints/checkpoint-2572/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:867d7aaa26f6e3569b6f9003dc4fee43956d36e10904f324cad06a3fbad98a66
+oid sha256:cfd5c7601ae95c77678dbcdbd11b24c0eaf41435cecfc1a415c0e302934a1790
 size 14244

training_checkpoints/checkpoint-2572/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d175d0495ec2005901bca7ae297edc36e69c6dfe66c46def57a73250c1f6823
+oid sha256:99918e207b19aaa32a525127eabf1c25da26aa315bfed3f2135ce7a1c9f91265
 size 1064

training_checkpoints/checkpoint-2572/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.14496386130680605,
+  "best_metric": 0.544445143585482,
   "best_model_checkpoint": "roberta-base/checkpoint-2572",
   "epoch": 4.0,
   "eval_steps": 500,
@@ -10,125 +10,125 @@
   "log_history": [
     {
       "epoch": 0.78,
-      "grad_norm": 6.534026145935059,
+      "grad_norm": 16.478424072265625,
       "learning_rate": 4.742871954380509e-05,
-      "loss": 2.2214,
+      "loss": 1.4921,
       "step": 500
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.24554608830364058,
-      "eval_f1_macro": 0.02628524046434494,
-      "eval_f1_micro": 0.24554608830364058,
-      "eval_f1_weighted": 0.09681356964210704,
-      "eval_loss": 2.1982107162475586,
-      "eval_macro_fpr": 0.06666666666666667,
-      "eval_macro_sensitivity": 0.06666666666666667,
-      "eval_macro_specificity": 0.9333333333333333,
-      "eval_precision": 0.06029288148121926,
-      "eval_precision_macro": 0.016369739220242704,
-      "eval_recall": 0.24554608830364058,
-      "eval_recall_macro": 0.06666666666666667,
-      "eval_runtime": 29.4439,
-      "eval_samples_per_second": 43.846,
-      "eval_steps_per_second": 5.502,
-      "eval_weighted_fpr": 0.17997043606799704,
-      "eval_weighted_sensitivity": 0.24554608830364058,
-      "eval_weighted_specificity": 0.7544539116963594,
+      "eval_accuracy": 0.6762199845081333,
+      "eval_f1_macro": 0.36762215914350266,
+      "eval_f1_micro": 0.6762199845081333,
+      "eval_f1_weighted": 0.6381553142782734,
+      "eval_loss": 1.0307424068450928,
+      "eval_macro_fpr": 0.03508857639652692,
+      "eval_macro_sensitivity": 0.4027183655220987,
+      "eval_macro_specificity": 0.9746655126680946,
+      "eval_precision": 0.6361951106730956,
+      "eval_precision_macro": 0.3745780279088578,
+      "eval_recall": 0.6762199845081333,
+      "eval_recall_macro": 0.4027183655220987,
+      "eval_runtime": 65.0146,
+      "eval_samples_per_second": 19.857,
+      "eval_steps_per_second": 2.492,
+      "eval_weighted_fpr": 0.03306962025316456,
+      "eval_weighted_sensitivity": 0.6762199845081333,
+      "eval_weighted_specificity": 0.9437627055132857,
       "step": 643
     },
     {
       "epoch": 1.56,
-      "grad_norm": 6.595059871673584,
+      "grad_norm": 11.052178382873535,
       "learning_rate": 4.483670295489891e-05,
-      "loss": 2.1462,
+      "loss": 0.9019,
       "step": 1000
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.3361735089078234,
-      "eval_f1_macro": 0.06552880630375763,
-      "eval_f1_micro": 0.3361735089078234,
-      "eval_f1_weighted": 0.2010480796875688,
-      "eval_loss": 1.9634978771209717,
-      "eval_macro_fpr": 0.09092506634165738,
-      "eval_macro_sensitivity": 0.11547020741602194,
-      "eval_macro_specificity": 0.9457066851291582,
-      "eval_precision": 0.1591168191834871,
-      "eval_precision_macro": 0.05308920125936356,
-      "eval_recall": 0.3361735089078234,
-      "eval_recall_macro": 0.11547020741602194,
-      "eval_runtime": 57.2389,
-      "eval_samples_per_second": 22.555,
-      "eval_steps_per_second": 2.83,
-      "eval_weighted_fpr": 0.12361171210154334,
-      "eval_weighted_sensitivity": 0.3361735089078234,
-      "eval_weighted_specificity": 0.84942676802955,
+      "eval_accuracy": 0.737412858249419,
+      "eval_f1_macro": 0.4767493392755858,
+      "eval_f1_micro": 0.737412858249419,
+      "eval_f1_weighted": 0.7197263095885438,
+      "eval_loss": 0.8135831952095032,
+      "eval_macro_fpr": 0.025093207469991694,
+      "eval_macro_sensitivity": 0.5148741493607558,
+      "eval_macro_specificity": 0.9804978566683986,
+      "eval_precision": 0.7158327481302543,
+      "eval_precision_macro": 0.4625056436667857,
+      "eval_recall": 0.737412858249419,
+      "eval_recall_macro": 0.5148741493607558,
+      "eval_runtime": 78.3833,
+      "eval_samples_per_second": 16.47,
+      "eval_steps_per_second": 2.067,
+      "eval_weighted_fpr": 0.024804273066510573,
+      "eval_weighted_sensitivity": 0.737412858249419,
+      "eval_weighted_specificity": 0.9700549917765612,
       "step": 1286
     },
     {
       "epoch": 2.33,
-      "grad_norm": 41.05898666381836,
-      "learning_rate": 4.224987039917055e-05,
-      "loss": 1.995,
+      "grad_norm": 16.875791549682617,
+      "learning_rate": 4.224468636599274e-05,
+      "loss": 0.7714,
       "step": 1500
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.3555383423702556,
-      "eval_f1_macro": 0.09312335075016234,
-      "eval_f1_micro": 0.3555383423702556,
-      "eval_f1_weighted": 0.2541975590745503,
-      "eval_loss": 1.932247281074524,
-      "eval_macro_fpr": 0.09537141294157879,
-      "eval_macro_sensitivity": 0.14351969127084646,
-      "eval_macro_specificity": 0.9497795632717652,
-      "eval_precision": 0.23412920314698393,
-      "eval_precision_macro": 0.09260367628400626,
-      "eval_recall": 0.3555383423702556,
-      "eval_recall_macro": 0.14351969127084646,
-      "eval_runtime": 59.8887,
-      "eval_samples_per_second": 21.557,
-      "eval_steps_per_second": 2.705,
-      "eval_weighted_fpr": 0.11463213006337834,
-      "eval_weighted_sensitivity": 0.3555383423702556,
-      "eval_weighted_specificity": 0.8911551067062222,
+      "eval_accuracy": 0.7621998450813323,
+      "eval_f1_macro": 0.5064301619528891,
+      "eval_f1_micro": 0.7621998450813323,
+      "eval_f1_weighted": 0.7440787580787277,
+      "eval_loss": 0.8493772149085999,
+      "eval_macro_fpr": 0.022439395825223803,
+      "eval_macro_sensitivity": 0.5353612181578932,
+      "eval_macro_specificity": 0.9822327546765126,
+      "eval_precision": 0.7385639759374297,
+      "eval_precision_macro": 0.505572707108006,
+      "eval_recall": 0.7621998450813323,
+      "eval_recall_macro": 0.5353612181578932,
+      "eval_runtime": 62.8149,
+      "eval_samples_per_second": 20.552,
+      "eval_steps_per_second": 2.579,
+      "eval_weighted_fpr": 0.02179933252858056,
+      "eval_weighted_sensitivity": 0.7621998450813323,
+      "eval_weighted_specificity": 0.9712914750663563,
       "step": 1929
     },
     {
       "epoch": 3.11,
-      "grad_norm": 4.287881851196289,
-      "learning_rate": 3.96630378434422e-05,
-      "loss": 2.0064,
+      "grad_norm": 59.48842239379883,
+      "learning_rate": 3.9652669777086575e-05,
+      "loss": 0.7018,
       "step": 2000
     },
     {
       "epoch": 3.89,
-      "grad_norm": 5.713494777679443,
-      "learning_rate": 3.7076205287713846e-05,
-      "loss": 1.8909,
+      "grad_norm": 1.6855363845825195,
+      "learning_rate": 3.706583722135822e-05,
+      "loss": 0.5351,
       "step": 2500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.3787761425251743,
-      "eval_f1_macro": 0.14496386130680605,
-      "eval_f1_micro": 0.3787761425251743,
-      "eval_f1_weighted": 0.29773714484044667,
-      "eval_loss": 1.8874640464782715,
-      "eval_macro_fpr": 0.08706956494211866,
-      "eval_macro_sensitivity": 0.2191470222337379,
-      "eval_macro_specificity": 0.9525422186332462,
-      "eval_precision": 0.28749388015966415,
-      "eval_precision_macro": 0.12703293957809556,
-      "eval_recall": 0.3787761425251743,
-      "eval_recall_macro": 0.2191470222337379,
-      "eval_runtime": 69.4305,
-      "eval_samples_per_second": 18.594,
-      "eval_steps_per_second": 2.333,
-      "eval_weighted_fpr": 0.10486401673640168,
-      "eval_weighted_sensitivity": 0.3787761425251743,
-      "eval_weighted_specificity": 0.9093571369735185,
+      "eval_accuracy": 0.7645236250968241,
+      "eval_f1_macro": 0.544445143585482,
+      "eval_f1_micro": 0.7645236250968241,
+      "eval_f1_weighted": 0.7545581478665218,
+      "eval_loss": 0.9904704093933105,
+      "eval_macro_fpr": 0.022368675533492623,
+      "eval_macro_sensitivity": 0.5653277909563001,
+      "eval_macro_specificity": 0.9821237739107975,
+      "eval_precision": 0.7656722240785093,
+      "eval_precision_macro": 0.5495765557734748,
+      "eval_recall": 0.7645236250968241,
+      "eval_recall_macro": 0.5653277909563001,
+      "eval_runtime": 80.7211,
+      "eval_samples_per_second": 15.993,
+      "eval_steps_per_second": 2.007,
+      "eval_weighted_fpr": 0.021526695935419914,
+      "eval_weighted_sensitivity": 0.7645236250968241,
+      "eval_weighted_specificity": 0.9673329835651372,
       "step": 2572
     }
   ],

training_checkpoints/checkpoint-2572/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6da235c4e1e89b590bb6c4dd543edc36c67b04a23baad1858c6c64a331ed4bd1
+oid sha256:beb8f128167a5d05587c32e6e3bff806f690d3ab0996792b1d08193bcb3299f2
 size 4920