updated model weights

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/mnt/beegfs/mc000051/CERPLES/Models/bert-base-spanish-wwm-cased",
+  "_name_or_path": "Models/bert-base-spanish-wwm-cased",
   "architectures": [
     "BertForSequenceClassification"
   ],

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d31a759ec106786b7d14b9db5903a8d1fc4094c5aba3dfa09c784b33bcb11b8
+oid sha256:6d0f5d55679c9649feef087aac6e35c24e2ba7740b6451422c9da4c0b3fe0e15
 size 879021981

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd2df4b7c392b8d623f36a4e2ef75af9a85e57170bab256fcca5ab95d80e9cf7
+oid sha256:14a497b1de3b9690a8def05e6522dc138ae0d30fe5d8b23d86922d491a3c206c
 size 439523757

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f01fef0c59d4860e25106e718f35edb1ef106ebaa4b62dfd95f8e7e4cf4e146e
+oid sha256:a1c5ffa6dfdfce8b9b2e8b1186fa1255461514df45bfa6889a547ccb5f4f00d9
 size 14567

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:392a982a9ac8e9d0e956548d877a9d8bcc8f61fbe896d0a05495f841534dd993
+oid sha256:bf40301a5f58872f87f1d2e7590a284756e863760c4895c4f9a55b272e36c722
 size 623

tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

tokenizer_config.json

@@ -1 +1 @@
-{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": false, "name_or_path": "/mnt/beegfs/mc000051/CERPLES/Models/bert-base-spanish-wwm-cased", "do_basic_tokenize": true, "never_split": null, "model_max_length": 512, "special_tokens_map_file": "/mnt/beegfs/mc000051/CERPLES/Models/bert-base-spanish-wwm-cased/special_tokens_map.json", "tokenizer_class": "BertTokenizer"}
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": false, "name_or_path": "Models/bert-base-spanish-wwm-cased", "do_basic_tokenize": true, "never_split": null, "model_max_length": 512, "special_tokens_map_file": "Models/bert-base-spanish-wwm-cased/special_tokens_map.json", "tokenizer_class": "BertTokenizer"}

trainer_state.json

@@ -1,156 +1,340 @@
 {
-  "best_metric": 0.6235998688300035,
-  "best_model_checkpoint": "./CARES/checkpoints/bert-bas/checkpoint-3500",
-  "epoch": 24.822695035460992,
-  "global_step": 3500,
+  "best_metric": 0.58822980593355,
+  "best_model_checkpoint": "./CARES/checkpoints/bert-ba/checkpoint-2961",
+  "epoch": 21.0,
+  "global_step": 2961,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.22213496267795563,
+      "eval_macro_f1": 0.08913506250646834,
+      "eval_macro_precision": 0.1601542788476893,
+      "eval_macro_recall": 0.07453283930073998,
+      "eval_micro_f1": 0.3941888619854721,
+      "eval_micro_precision": 0.7650375939849624,
+      "eval_micro_recall": 0.2654924983692107,
+      "eval_runtime": 3.0004,
+      "eval_samples_per_second": 321.956,
+      "eval_steps_per_second": 20.331,
+      "step": 141
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.18322525918483734,
+      "eval_macro_f1": 0.20542764030139582,
+      "eval_macro_precision": 0.31153358140506404,
+      "eval_macro_recall": 0.1749941105591186,
+      "eval_micro_f1": 0.6042759177087536,
+      "eval_micro_precision": 0.7917547568710359,
+      "eval_micro_recall": 0.4885844748858447,
+      "eval_runtime": 3.0066,
+      "eval_samples_per_second": 321.295,
+      "eval_steps_per_second": 20.289,
+      "step": 282
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.168878972530365,
+      "eval_macro_f1": 0.29169004336353355,
+      "eval_macro_precision": 0.39228947650947077,
+      "eval_macro_recall": 0.2624222179377449,
+      "eval_micro_f1": 0.6591789310611929,
+      "eval_micro_precision": 0.8112488083889419,
+      "eval_micro_recall": 0.5551206784083497,
+      "eval_runtime": 3.0155,
+      "eval_samples_per_second": 320.35,
+      "eval_steps_per_second": 20.229,
+      "step": 423
+    },
     {
       "epoch": 3.55,
       "learning_rate": 2.8936170212765956e-05,
-      "loss": 0.2014,
+      "loss": 0.1959,
       "step": 500
     },
     {
-      "epoch": 3.55,
-      "eval_loss": 0.15534088015556335,
-      "eval_macro_f1": 0.34191398087233876,
-      "eval_macro_precision": 0.4575725498723576,
-      "eval_macro_recall": 0.32634682637580636,
-      "eval_micro_f1": 0.7120380534211489,
-      "eval_micro_precision": 0.8108333333333333,
-      "eval_micro_recall": 0.634703196347032,
-      "eval_runtime": 2.9867,
-      "eval_samples_per_second": 323.432,
-      "eval_steps_per_second": 20.424,
-      "step": 500
+      "epoch": 4.0,
+      "eval_loss": 0.1501942127943039,
+      "eval_macro_f1": 0.35589775613243146,
+      "eval_macro_precision": 0.4667524057986162,
+      "eval_macro_recall": 0.3166674518971492,
+      "eval_micro_f1": 0.7155012948575658,
+      "eval_micro_precision": 0.8264957264957264,
+      "eval_micro_recall": 0.6307893020221788,
+      "eval_runtime": 3.0213,
+      "eval_samples_per_second": 319.733,
+      "eval_steps_per_second": 20.19,
+      "step": 564
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.14672552049160004,
+      "eval_macro_f1": 0.47974247836953543,
+      "eval_macro_precision": 0.6466753641860046,
+      "eval_macro_recall": 0.42420722655395326,
+      "eval_micro_f1": 0.7180585296216989,
+      "eval_micro_precision": 0.7927501970055162,
+      "eval_micro_recall": 0.6562296151337247,
+      "eval_runtime": 3.0023,
+      "eval_samples_per_second": 321.752,
+      "eval_steps_per_second": 20.318,
+      "step": 705
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.14830899238586426,
+      "eval_macro_f1": 0.4652554310103303,
+      "eval_macro_precision": 0.6251017145329125,
+      "eval_macro_recall": 0.4150412963066584,
+      "eval_micro_f1": 0.7256140350877194,
+      "eval_micro_precision": 0.7851176917236142,
+      "eval_micro_recall": 0.6744944553163731,
+      "eval_runtime": 2.9724,
+      "eval_samples_per_second": 324.991,
+      "eval_steps_per_second": 20.522,
+      "step": 846
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.1467733532190323,
+      "eval_macro_f1": 0.5097890383483837,
+      "eval_macro_precision": 0.6555313836604214,
+      "eval_macro_recall": 0.4437744774893355,
+      "eval_micro_f1": 0.7510431154381085,
+      "eval_micro_precision": 0.8041697691734921,
+      "eval_micro_recall": 0.7045009784735812,
+      "eval_runtime": 2.9801,
+      "eval_samples_per_second": 324.15,
+      "eval_steps_per_second": 20.469,
+      "step": 987
     },
     {
       "epoch": 7.09,
       "learning_rate": 2.7872340425531914e-05,
-      "loss": 0.0909,
+      "loss": 0.0904,
       "step": 1000
     },
     {
-      "epoch": 7.09,
-      "eval_loss": 0.14328321814537048,
-      "eval_macro_f1": 0.5216600519885811,
-      "eval_macro_precision": 0.6582415980876382,
-      "eval_macro_recall": 0.45829582028075544,
-      "eval_micro_f1": 0.7527932960893854,
-      "eval_micro_precision": 0.8099173553719008,
-      "eval_micro_recall": 0.7031963470319634,
-      "eval_runtime": 2.9852,
-      "eval_samples_per_second": 323.592,
-      "eval_steps_per_second": 20.434,
-      "step": 1000
+      "epoch": 8.0,
+      "eval_loss": 0.14951790869235992,
+      "eval_macro_f1": 0.5411089496886734,
+      "eval_macro_precision": 0.6577005724429985,
+      "eval_macro_recall": 0.47852188216570773,
+      "eval_micro_f1": 0.7489539748953974,
+      "eval_micro_precision": 0.8044943820224719,
+      "eval_micro_recall": 0.700587084148728,
+      "eval_runtime": 2.9774,
+      "eval_samples_per_second": 324.441,
+      "eval_steps_per_second": 20.487,
+      "step": 1128
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 0.1488533467054367,
+      "eval_macro_f1": 0.5590641407878995,
+      "eval_macro_precision": 0.6102853068447548,
+      "eval_macro_recall": 0.5269868274318028,
+      "eval_micro_f1": 0.755420054200542,
+      "eval_micro_precision": 0.7857646229739254,
+      "eval_micro_recall": 0.7273320287018917,
+      "eval_runtime": 2.9723,
+      "eval_samples_per_second": 324.996,
+      "eval_steps_per_second": 20.523,
+      "step": 1269
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 0.15144900977611542,
+      "eval_macro_f1": 0.5597430867412742,
+      "eval_macro_precision": 0.6112400331236583,
+      "eval_macro_recall": 0.5287805393050224,
+      "eval_micro_f1": 0.7608550434201737,
+      "eval_micro_precision": 0.7796030116358659,
+      "eval_micro_recall": 0.7429876060013046,
+      "eval_runtime": 2.9735,
+      "eval_samples_per_second": 324.869,
+      "eval_steps_per_second": 20.515,
+      "step": 1410
     },
     {
       "epoch": 10.64,
       "learning_rate": 2.6808510638297873e-05,
-      "loss": 0.0428,
+      "loss": 0.0424,
       "step": 1500
     },
     {
-      "epoch": 10.64,
-      "eval_loss": 0.1534910351037979,
-      "eval_macro_f1": 0.5762696837727688,
-      "eval_macro_precision": 0.6633765574994241,
-      "eval_macro_recall": 0.5217654870639399,
-      "eval_micro_f1": 0.7763426240652617,
-      "eval_micro_precision": 0.8105039034776437,
-      "eval_micro_recall": 0.7449445531637312,
-      "eval_runtime": 2.9874,
-      "eval_samples_per_second": 323.358,
-      "eval_steps_per_second": 20.419,
-      "step": 1500
+      "epoch": 11.0,
+      "eval_loss": 0.15306253731250763,
+      "eval_macro_f1": 0.5476291897617486,
+      "eval_macro_precision": 0.6042943362522458,
+      "eval_macro_recall": 0.5150958586698227,
+      "eval_micro_f1": 0.7621293800539084,
+      "eval_micro_precision": 0.7881533101045296,
+      "eval_micro_recall": 0.7377690802348337,
+      "eval_runtime": 2.9994,
+      "eval_samples_per_second": 322.068,
+      "eval_steps_per_second": 20.338,
+      "step": 1551
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 0.15226905047893524,
+      "eval_macro_f1": 0.5645136859707334,
+      "eval_macro_precision": 0.6250404048441249,
+      "eval_macro_recall": 0.5290350408209084,
+      "eval_micro_f1": 0.7750586657727119,
+      "eval_micro_precision": 0.7972413793103448,
+      "eval_micro_recall": 0.7540769732550554,
+      "eval_runtime": 2.9853,
+      "eval_samples_per_second": 323.581,
+      "eval_steps_per_second": 20.433,
+      "step": 1692
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 0.15537378191947937,
+      "eval_macro_f1": 0.5801357636140765,
+      "eval_macro_precision": 0.6557226559864182,
+      "eval_macro_recall": 0.5300773127347133,
+      "eval_micro_f1": 0.7715736040609137,
+      "eval_micro_precision": 0.8016877637130801,
+      "eval_micro_recall": 0.7436399217221135,
+      "eval_runtime": 4.9377,
+      "eval_samples_per_second": 195.637,
+      "eval_steps_per_second": 12.354,
+      "step": 1833
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 0.15421901643276215,
+      "eval_macro_f1": 0.5773329417808055,
+      "eval_macro_precision": 0.6485175319800287,
+      "eval_macro_recall": 0.5350306451399538,
+      "eval_micro_f1": 0.77552400270453,
+      "eval_micro_precision": 0.8049122807017544,
+      "eval_micro_recall": 0.7482061317677756,
+      "eval_runtime": 3.0097,
+      "eval_samples_per_second": 320.96,
+      "eval_steps_per_second": 20.268,
+      "step": 1974
     },
     {
       "epoch": 14.18,
       "learning_rate": 2.574468085106383e-05,
-      "loss": 0.0231,
+      "loss": 0.0219,
       "step": 2000
     },
     {
-      "epoch": 14.18,
-      "eval_loss": 0.15115897357463837,
-      "eval_macro_f1": 0.5809807709622608,
-      "eval_macro_precision": 0.6166814111331129,
-      "eval_macro_recall": 0.5608937720056061,
-      "eval_micro_f1": 0.7775925308436145,
-      "eval_micro_precision": 0.7953615279672579,
-      "eval_micro_recall": 0.7606001304631441,
-      "eval_runtime": 2.985,
-      "eval_samples_per_second": 323.621,
-      "eval_steps_per_second": 20.436,
-      "step": 2000
+      "epoch": 15.0,
+      "eval_loss": 0.15735264122486115,
+      "eval_macro_f1": 0.5661412601117408,
+      "eval_macro_precision": 0.5937031922346153,
+      "eval_macro_recall": 0.5467428221378843,
+      "eval_micro_f1": 0.7738255033557048,
+      "eval_micro_precision": 0.796821008984105,
+      "eval_micro_recall": 0.7521200260926288,
+      "eval_runtime": 2.9854,
+      "eval_samples_per_second": 323.576,
+      "eval_steps_per_second": 20.433,
+      "step": 2115
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 0.1617126762866974,
+      "eval_macro_f1": 0.5686269653829785,
+      "eval_macro_precision": 0.6170850785135864,
+      "eval_macro_recall": 0.5368268361090598,
+      "eval_micro_f1": 0.779524924723988,
+      "eval_micro_precision": 0.8001373626373627,
+      "eval_micro_recall": 0.7599478147423353,
+      "eval_runtime": 2.9891,
+      "eval_samples_per_second": 323.178,
+      "eval_steps_per_second": 20.408,
+      "step": 2256
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 0.16439199447631836,
+      "eval_macro_f1": 0.5741609962540627,
+      "eval_macro_precision": 0.6203354824423237,
+      "eval_macro_recall": 0.5508776094197674,
+      "eval_micro_f1": 0.7765251989389921,
+      "eval_micro_precision": 0.7896156439649359,
+      "eval_micro_recall": 0.7638617090671885,
+      "eval_runtime": 2.979,
+      "eval_samples_per_second": 324.271,
+      "eval_steps_per_second": 20.477,
+      "step": 2397
     },
     {
       "epoch": 17.73,
       "learning_rate": 2.4680851063829786e-05,
-      "loss": 0.0137,
+      "loss": 0.013,
       "step": 2500
     },
     {
-      "epoch": 17.73,
-      "eval_loss": 0.16502933204174042,
-      "eval_macro_f1": 0.5991629446392395,
-      "eval_macro_precision": 0.6221331811310129,
-      "eval_macro_recall": 0.5815882875339234,
-      "eval_micro_f1": 0.7877187190491911,
-      "eval_micro_precision": 0.7974598930481284,
-      "eval_micro_recall": 0.7782126549249837,
-      "eval_runtime": 2.9854,
-      "eval_samples_per_second": 323.576,
-      "eval_steps_per_second": 20.433,
-      "step": 2500
+      "epoch": 18.0,
+      "eval_loss": 0.16467925906181335,
+      "eval_macro_f1": 0.5828461745953328,
+      "eval_macro_precision": 0.6374781102044266,
+      "eval_macro_recall": 0.5491770837420844,
+      "eval_micro_f1": 0.7797519275896748,
+      "eval_micro_precision": 0.8020689655172414,
+      "eval_micro_recall": 0.7586431833007176,
+      "eval_runtime": 2.9765,
+      "eval_samples_per_second": 324.537,
+      "eval_steps_per_second": 20.494,
+      "step": 2538
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 0.17651152610778809,
+      "eval_macro_f1": 0.578653206536447,
+      "eval_macro_precision": 0.5940445993831301,
+      "eval_macro_recall": 0.5738532439728641,
+      "eval_micro_f1": 0.7715868361029651,
+      "eval_micro_precision": 0.7708333333333334,
+      "eval_micro_recall": 0.7723418134377038,
+      "eval_runtime": 2.9833,
+      "eval_samples_per_second": 323.805,
+      "eval_steps_per_second": 20.447,
+      "step": 2679
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 0.17114990949630737,
+      "eval_macro_f1": 0.5779704862006727,
+      "eval_macro_precision": 0.6285916131953179,
+      "eval_macro_recall": 0.5455200342016094,
+      "eval_micro_f1": 0.7783711615487315,
+      "eval_micro_precision": 0.7969924812030075,
+      "eval_micro_recall": 0.7606001304631441,
+      "eval_runtime": 2.9774,
+      "eval_samples_per_second": 324.449,
+      "eval_steps_per_second": 20.488,
+      "step": 2820
     },
     {
-      "epoch": 21.28,
-      "learning_rate": 2.3617021276595744e-05,
-      "loss": 0.0087,
-      "step": 3000
-    },
-    {
-      "epoch": 21.28,
-      "eval_loss": 0.1669953465461731,
-      "eval_macro_f1": 0.6173151275498924,
-      "eval_macro_precision": 0.6814989884732782,
-      "eval_macro_recall": 0.5992296363093449,
-      "eval_micro_f1": 0.7861842105263159,
-      "eval_micro_precision": 0.7929661579296616,
-      "eval_micro_recall": 0.7795172863666014,
-      "eval_runtime": 2.9877,
-      "eval_samples_per_second": 323.321,
-      "eval_steps_per_second": 20.417,
-      "step": 3000
-    },
-    {
-      "epoch": 24.82,
-      "learning_rate": 2.2553191489361703e-05,
-      "loss": 0.0063,
-      "step": 3500
-    },
-    {
-      "epoch": 24.82,
-      "eval_loss": 0.18081951141357422,
-      "eval_macro_f1": 0.6235998688300035,
-      "eval_macro_precision": 0.6667425956441049,
-      "eval_macro_recall": 0.5993735670848622,
-      "eval_micro_f1": 0.7835120643431636,
-      "eval_micro_precision": 0.8056512749827704,
-      "eval_micro_recall": 0.7625570776255708,
-      "eval_runtime": 2.9803,
-      "eval_samples_per_second": 324.127,
-      "eval_steps_per_second": 20.468,
-      "step": 3500
+      "epoch": 21.0,
+      "eval_loss": 0.1727105975151062,
+      "eval_macro_f1": 0.58822980593355,
+      "eval_macro_precision": 0.6578174885933932,
+      "eval_macro_recall": 0.5467783967982416,
+      "eval_micro_f1": 0.782282793867121,
+      "eval_micro_precision": 0.818830242510699,
+      "eval_micro_recall": 0.7488584474885844,
+      "eval_runtime": 2.9806,
+      "eval_samples_per_second": 324.101,
+      "eval_steps_per_second": 20.466,
+      "step": 2961
     }
   ],
   "max_steps": 14100,
   "num_train_epochs": 100,
-  "total_flos": 1.4624642507081568e+16,
+  "total_flos": 1.2360197339395008e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b32d80d5d5dce67c9a2a11756aa0d57c15b20ca460c1ef2d87c1f7b006d31804
-size 3247
+oid sha256:438b946a390e0a080e0951e1d1cf250efa3125664bf79c3baf90d4c23e353fcb
+size 3183