End of training

README.md

@@ -4,7 +4,7 @@ base_model: xlm-roberta-base
 tags:
 - generated_from_trainer
 datasets:
-- universal_ner
+- universalner/universal_ner
 metrics:
 - precision
 - recall
@@ -17,8 +17,8 @@ model-index:
       name: Token Classification
       type: token-classification
     dataset:
-      name: universal_ner
-      type: universal_ner
+      name: universalner/universal_ner en_ewt
+      type: universalner/universal_ner
       config: en_ewt
       split: validation
       args: en_ewt
@@ -42,7 +42,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # UNER_subword_tk_en_lora_alpha_64_drop_0.3_rank_32_seed_42
 
-This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the universal_ner dataset.
+This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the universalner/universal_ner en_ewt dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.0607
 - Precision: 0.7736

all_results.json

@@ -1,18 +1,18 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.9841743210465624,
-    "eval_f1": 0.8008028098344204,
-    "eval_loss": 0.06448859721422195,
-    "eval_precision": 0.7770204479065238,
-    "eval_recall": 0.8260869565217391,
-    "eval_runtime": 6.7213,
+    "eval_accuracy": 0.9840550320092251,
+    "eval_f1": 0.7979949874686717,
+    "eval_loss": 0.060650501400232315,
+    "eval_precision": 0.7735665694849369,
+    "eval_recall": 0.8240165631469979,
+    "eval_runtime": 2.1367,
     "eval_samples": 2001,
-    "eval_samples_per_second": 297.711,
-    "eval_steps_per_second": 37.344,
+    "eval_samples_per_second": 936.473,
+    "eval_steps_per_second": 29.484,
     "total_flos": 8817316923832968.0,
     "train_loss": 0.038693137679781234,
-    "train_runtime": 671.6771,
+    "train_runtime": 596.8038,
     "train_samples": 12543,
-    "train_samples_per_second": 373.483,
-    "train_steps_per_second": 11.672
+    "train_samples_per_second": 420.339,
+    "train_steps_per_second": 13.137
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.9841743210465624,
-    "eval_f1": 0.8008028098344204,
-    "eval_loss": 0.06448859721422195,
-    "eval_precision": 0.7770204479065238,
-    "eval_recall": 0.8260869565217391,
-    "eval_runtime": 6.7213,
+    "eval_accuracy": 0.9840550320092251,
+    "eval_f1": 0.7979949874686717,
+    "eval_loss": 0.060650501400232315,
+    "eval_precision": 0.7735665694849369,
+    "eval_recall": 0.8240165631469979,
+    "eval_runtime": 2.1367,
     "eval_samples": 2001,
-    "eval_samples_per_second": 297.711,
-    "eval_steps_per_second": 37.344
+    "eval_samples_per_second": 936.473,
+    "eval_steps_per_second": 29.484
 }

train_results.json

@@ -2,8 +2,8 @@
     "epoch": 20.0,
     "total_flos": 8817316923832968.0,
     "train_loss": 0.038693137679781234,
-    "train_runtime": 671.6771,
+    "train_runtime": 596.8038,
     "train_samples": 12543,
-    "train_samples_per_second": 373.483,
-    "train_steps_per_second": 11.672
+    "train_samples_per_second": 420.339,
+    "train_steps_per_second": 13.137
 }

trainer_state.json

@@ -10,14 +10,14 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9740745158853235,
-      "eval_f1": 0.6371191135734072,
-      "eval_loss": 0.09594480693340302,
-      "eval_precision": 0.575,
+      "eval_accuracy": 0.973955226847986,
+      "eval_f1": 0.6374133949191687,
+      "eval_loss": 0.08990096300840378,
+      "eval_precision": 0.5754795663052544,
       "eval_recall": 0.7142857142857143,
-      "eval_runtime": 6.7658,
-      "eval_samples_per_second": 295.753,
-      "eval_steps_per_second": 37.099,
+      "eval_runtime": 1.9571,
+      "eval_samples_per_second": 1022.422,
+      "eval_steps_per_second": 32.19,
       "step": 392
     },
     {
@@ -29,14 +29,14 @@
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9798401526899678,
+      "eval_accuracy": 0.9798799157024136,
       "eval_f1": 0.7409200968523002,
-      "eval_loss": 0.07035845518112183,
+      "eval_loss": 0.06505557149648666,
       "eval_precision": 0.6960873521383075,
       "eval_recall": 0.7919254658385093,
-      "eval_runtime": 6.6994,
-      "eval_samples_per_second": 298.683,
-      "eval_steps_per_second": 37.466,
+      "eval_runtime": 1.9464,
+      "eval_samples_per_second": 1028.043,
+      "eval_steps_per_second": 32.367,
       "step": 784
     },
     {
@@ -48,14 +48,14 @@
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9814306731878008,
+      "eval_accuracy": 0.9814704362002465,
       "eval_f1": 0.7639691714836224,
-      "eval_loss": 0.07102187722921371,
+      "eval_loss": 0.06638560444116592,
       "eval_precision": 0.7144144144144144,
       "eval_recall": 0.8209109730848861,
-      "eval_runtime": 6.7074,
-      "eval_samples_per_second": 298.328,
-      "eval_steps_per_second": 37.421,
+      "eval_runtime": 1.9458,
+      "eval_samples_per_second": 1028.343,
+      "eval_steps_per_second": 32.377,
       "step": 1176
     },
     {
@@ -67,26 +67,26 @@
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9820668813869339,
+      "eval_accuracy": 0.9821464074118256,
       "eval_f1": 0.7654440154440154,
-      "eval_loss": 0.06226913630962372,
+      "eval_loss": 0.05806884169578552,
       "eval_precision": 0.716998191681736,
       "eval_recall": 0.8209109730848861,
-      "eval_runtime": 6.6379,
-      "eval_samples_per_second": 301.45,
-      "eval_steps_per_second": 37.813,
+      "eval_runtime": 1.9555,
+      "eval_samples_per_second": 1023.292,
+      "eval_steps_per_second": 32.218,
       "step": 1568
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.9819873553620422,
       "eval_f1": 0.7762645914396887,
-      "eval_loss": 0.06265858560800552,
+      "eval_loss": 0.05841425433754921,
       "eval_precision": 0.7321100917431193,
       "eval_recall": 0.8260869565217391,
-      "eval_runtime": 6.7117,
-      "eval_samples_per_second": 298.137,
-      "eval_steps_per_second": 37.398,
+      "eval_runtime": 1.9706,
+      "eval_samples_per_second": 1015.404,
+      "eval_steps_per_second": 31.969,
       "step": 1960
     },
     {
@@ -98,14 +98,14 @@
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9826235635611754,
-      "eval_f1": 0.7796278158667973,
-      "eval_loss": 0.06144500896334648,
-      "eval_precision": 0.7397769516728625,
-      "eval_recall": 0.8240165631469979,
-      "eval_runtime": 6.6293,
-      "eval_samples_per_second": 301.841,
-      "eval_steps_per_second": 37.862,
+      "eval_accuracy": 0.9826633265736212,
+      "eval_f1": 0.7797940166748406,
+      "eval_loss": 0.05712873861193657,
+      "eval_precision": 0.7409133271202236,
+      "eval_recall": 0.8229813664596274,
+      "eval_runtime": 1.977,
+      "eval_samples_per_second": 1012.156,
+      "eval_steps_per_second": 31.867,
       "step": 2352
     },
     {
@@ -117,14 +117,14 @@
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9828621416358503,
+      "eval_accuracy": 0.9829019046482962,
       "eval_f1": 0.7857491459248416,
-      "eval_loss": 0.06205696985125542,
+      "eval_loss": 0.05804689601063728,
       "eval_precision": 0.7433056325023084,
       "eval_recall": 0.8333333333333334,
-      "eval_runtime": 6.7173,
-      "eval_samples_per_second": 297.888,
-      "eval_steps_per_second": 37.366,
+      "eval_runtime": 1.9531,
+      "eval_samples_per_second": 1024.51,
+      "eval_steps_per_second": 32.256,
       "step": 2744
     },
     {
@@ -136,14 +136,14 @@
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9846514771959124,
-      "eval_f1": 0.8008008008008008,
-      "eval_loss": 0.06194039061665535,
-      "eval_precision": 0.7751937984496124,
+      "eval_accuracy": 0.9846117141834665,
+      "eval_f1": 0.8004002001000501,
+      "eval_loss": 0.05779951810836792,
+      "eval_precision": 0.7744433688286544,
       "eval_recall": 0.8281573498964804,
-      "eval_runtime": 6.6418,
-      "eval_samples_per_second": 301.272,
-      "eval_steps_per_second": 37.791,
+      "eval_runtime": 1.9641,
+      "eval_samples_per_second": 1018.804,
+      "eval_steps_per_second": 32.076,
       "step": 3136
     },
     {
@@ -155,26 +155,26 @@
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.9834188238100918,
-      "eval_f1": 0.7904903417533433,
-      "eval_loss": 0.06049573794007301,
-      "eval_precision": 0.7578347578347578,
-      "eval_recall": 0.8260869565217391,
-      "eval_runtime": 6.7155,
-      "eval_samples_per_second": 297.969,
-      "eval_steps_per_second": 37.376,
+      "eval_accuracy": 0.9834585868225377,
+      "eval_f1": 0.7914809311540367,
+      "eval_loss": 0.056642863899469376,
+      "eval_precision": 0.7587844254510921,
+      "eval_recall": 0.8271221532091098,
+      "eval_runtime": 1.9436,
+      "eval_samples_per_second": 1029.513,
+      "eval_steps_per_second": 32.413,
       "step": 3528
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.9848502922581415,
-      "eval_f1": 0.804,
-      "eval_loss": 0.05974126234650612,
-      "eval_precision": 0.7775628626692457,
-      "eval_recall": 0.8322981366459627,
-      "eval_runtime": 6.7117,
-      "eval_samples_per_second": 298.137,
-      "eval_steps_per_second": 37.397,
+      "eval_accuracy": 0.9847707662332499,
+      "eval_f1": 0.8019999999999999,
+      "eval_loss": 0.05644238740205765,
+      "eval_precision": 0.7756286266924565,
+      "eval_recall": 0.8302277432712215,
+      "eval_runtime": 1.9568,
+      "eval_samples_per_second": 1022.58,
+      "eval_steps_per_second": 32.195,
       "step": 3920
     },
     {
@@ -187,13 +187,13 @@
     {
       "epoch": 11.0,
       "eval_accuracy": 0.9834983498349835,
-      "eval_f1": 0.7948717948717949,
-      "eval_loss": 0.05979320779442787,
-      "eval_precision": 0.7589453860640302,
+      "eval_f1": 0.7952639368524913,
+      "eval_loss": 0.05572296306490898,
+      "eval_precision": 0.7596606974552309,
       "eval_recall": 0.8343685300207039,
-      "eval_runtime": 6.6975,
-      "eval_samples_per_second": 298.769,
-      "eval_steps_per_second": 37.477,
+      "eval_runtime": 1.9713,
+      "eval_samples_per_second": 1015.086,
+      "eval_steps_per_second": 31.959,
       "step": 4312
     },
     {
@@ -205,14 +205,14 @@
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.9841743210465624,
-      "eval_f1": 0.798183652875883,
-      "eval_loss": 0.06226903572678566,
-      "eval_precision": 0.7785433070866141,
-      "eval_recall": 0.8188405797101449,
-      "eval_runtime": 6.6365,
-      "eval_samples_per_second": 301.514,
-      "eval_steps_per_second": 37.821,
+      "eval_accuracy": 0.9840550320092251,
+      "eval_f1": 0.7967644084934277,
+      "eval_loss": 0.05849658325314522,
+      "eval_precision": 0.7786561264822134,
+      "eval_recall": 0.8157349896480331,
+      "eval_runtime": 1.9705,
+      "eval_samples_per_second": 1015.502,
+      "eval_steps_per_second": 31.972,
       "step": 4704
     },
     {
@@ -224,26 +224,26 @@
     },
     {
       "epoch": 13.0,
-      "eval_accuracy": 0.9844924251461291,
-      "eval_f1": 0.8012078510317061,
-      "eval_loss": 0.06113631650805473,
-      "eval_precision": 0.7796278158667973,
+      "eval_accuracy": 0.9845321881585749,
+      "eval_f1": 0.802015113350126,
+      "eval_loss": 0.05740697309374809,
+      "eval_precision": 0.7811579980372915,
       "eval_recall": 0.8240165631469979,
-      "eval_runtime": 6.6332,
-      "eval_samples_per_second": 301.666,
-      "eval_steps_per_second": 37.84,
+      "eval_runtime": 1.9671,
+      "eval_samples_per_second": 1017.255,
+      "eval_steps_per_second": 32.028,
       "step": 5096
     },
     {
       "epoch": 14.0,
-      "eval_accuracy": 0.9835381128474293,
-      "eval_f1": 0.7942772570300938,
-      "eval_loss": 0.06113998964428902,
-      "eval_precision": 0.7587181903864278,
-      "eval_recall": 0.8333333333333334,
-      "eval_runtime": 6.6539,
-      "eval_samples_per_second": 300.725,
-      "eval_steps_per_second": 37.722,
+      "eval_accuracy": 0.983617638872321,
+      "eval_f1": 0.7956564659427443,
+      "eval_loss": 0.0576859749853611,
+      "eval_precision": 0.7603773584905661,
+      "eval_recall": 0.8343685300207039,
+      "eval_runtime": 1.9779,
+      "eval_samples_per_second": 1011.701,
+      "eval_steps_per_second": 31.853,
       "step": 5488
     },
     {
@@ -255,14 +255,14 @@
     },
     {
       "epoch": 15.0,
-      "eval_accuracy": 0.9844128991212374,
-      "eval_f1": 0.8028098344204717,
-      "eval_loss": 0.06239597126841545,
-      "eval_precision": 0.7789678675754625,
+      "eval_accuracy": 0.9845321881585749,
+      "eval_f1": 0.8040201005025126,
+      "eval_loss": 0.058342624455690384,
+      "eval_precision": 0.78125,
       "eval_recall": 0.8281573498964804,
-      "eval_runtime": 6.7198,
-      "eval_samples_per_second": 297.776,
-      "eval_steps_per_second": 37.352,
+      "eval_runtime": 1.9542,
+      "eval_samples_per_second": 1023.925,
+      "eval_steps_per_second": 32.238,
       "step": 5880
     },
     {
@@ -274,14 +274,14 @@
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.9841345580341166,
-      "eval_f1": 0.8019900497512437,
-      "eval_loss": 0.06371314078569412,
-      "eval_precision": 0.7720306513409961,
-      "eval_recall": 0.8343685300207039,
-      "eval_runtime": 6.6408,
-      "eval_samples_per_second": 301.319,
-      "eval_steps_per_second": 37.797,
+      "eval_accuracy": 0.9840947950216709,
+      "eval_f1": 0.802192326856004,
+      "eval_loss": 0.05951229855418205,
+      "eval_precision": 0.7732949087415946,
+      "eval_recall": 0.8333333333333334,
+      "eval_runtime": 1.9485,
+      "eval_samples_per_second": 1026.943,
+      "eval_steps_per_second": 32.333,
       "step": 6272
     },
     {
@@ -293,14 +293,14 @@
     },
     {
       "epoch": 17.0,
-      "eval_accuracy": 0.9839755059843334,
-      "eval_f1": 0.8014042126379137,
-      "eval_loss": 0.06424978375434875,
-      "eval_precision": 0.7772373540856031,
-      "eval_recall": 0.8271221532091098,
-      "eval_runtime": 6.6399,
-      "eval_samples_per_second": 301.361,
-      "eval_steps_per_second": 37.802,
+      "eval_accuracy": 0.9839357429718876,
+      "eval_f1": 0.8,
+      "eval_loss": 0.06028683856129646,
+      "eval_precision": 0.7755102040816326,
+      "eval_recall": 0.8260869565217391,
+      "eval_runtime": 1.9515,
+      "eval_samples_per_second": 1025.339,
+      "eval_steps_per_second": 32.282,
       "step": 6664
     },
     {
@@ -314,24 +314,24 @@
       "epoch": 18.0,
       "eval_accuracy": 0.9839755059843334,
       "eval_f1": 0.8004002001000501,
-      "eval_loss": 0.06440357118844986,
+      "eval_loss": 0.06049268692731857,
       "eval_precision": 0.7744433688286544,
       "eval_recall": 0.8281573498964804,
-      "eval_runtime": 6.7187,
-      "eval_samples_per_second": 297.827,
-      "eval_steps_per_second": 37.359,
+      "eval_runtime": 1.9474,
+      "eval_samples_per_second": 1027.524,
+      "eval_steps_per_second": 32.351,
       "step": 7056
     },
     {
       "epoch": 19.0,
-      "eval_accuracy": 0.9842538470714541,
-      "eval_f1": 0.8035982008995503,
-      "eval_loss": 0.06508098542690277,
-      "eval_precision": 0.7768115942028986,
-      "eval_recall": 0.8322981366459627,
-      "eval_runtime": 6.7259,
-      "eval_samples_per_second": 297.506,
-      "eval_steps_per_second": 37.318,
+      "eval_accuracy": 0.9841743210465624,
+      "eval_f1": 0.8041958041958043,
+      "eval_loss": 0.06108159199357033,
+      "eval_precision": 0.777027027027027,
+      "eval_recall": 0.8333333333333334,
+      "eval_runtime": 1.9732,
+      "eval_samples_per_second": 1014.065,
+      "eval_steps_per_second": 31.927,
       "step": 7448
     },
     {
@@ -343,14 +343,14 @@
     },
     {
       "epoch": 20.0,
-      "eval_accuracy": 0.9841743210465624,
-      "eval_f1": 0.8008028098344204,
-      "eval_loss": 0.06448859721422195,
-      "eval_precision": 0.7770204479065238,
-      "eval_recall": 0.8260869565217391,
-      "eval_runtime": 6.703,
-      "eval_samples_per_second": 298.525,
-      "eval_steps_per_second": 37.446,
+      "eval_accuracy": 0.9840550320092251,
+      "eval_f1": 0.7979949874686717,
+      "eval_loss": 0.060650501400232315,
+      "eval_precision": 0.7735665694849369,
+      "eval_recall": 0.8240165631469979,
+      "eval_runtime": 1.9653,
+      "eval_samples_per_second": 1018.178,
+      "eval_steps_per_second": 32.057,
       "step": 7840
     },
     {
@@ -358,9 +358,9 @@
       "step": 7840,
       "total_flos": 8817316923832968.0,
       "train_loss": 0.038693137679781234,
-      "train_runtime": 671.6771,
-      "train_samples_per_second": 373.483,
-      "train_steps_per_second": 11.672
+      "train_runtime": 596.8038,
+      "train_samples_per_second": 420.339,
+      "train_steps_per_second": 13.137
     }
   ],
   "logging_steps": 500,