End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
-    "train_loss": 0.18657105684280395,
-    "train_runtime": 93.1154,
+    "train_loss": 0.17726608804294042,
+    "train_runtime": 611.8323,
     "train_samples": 1120,
-    "train_samples_per_second": 60.14,
-    "train_steps_per_second": 3.759
+    "train_samples_per_second": 9.153,
+    "train_steps_per_second": 1.144
 }

predict_results_fake-news.txt

@@ -1,6 +1,6 @@
 index	prediction
 0	1
-1	1
+1	0
 2	1
 3	1
 4	1
@@ -35,13 +35,13 @@ index	prediction
 33	1
 34	1
 35	1
-36	1
+36	0
 37	1
 38	0
 39	0
 40	0
 41	1
-42	1
+42	0
 43	1
 44	0
 45	0
@@ -54,7 +54,7 @@ index	prediction
 52	0
 53	1
 54	1
-55	0
+55	1
 56	0
 57	1
 58	1
@@ -64,7 +64,7 @@ index	prediction
 62	1
 63	0
 64	1
-65	1
+65	0
 66	1
 67	0
 68	0
@@ -75,8 +75,8 @@ index	prediction
 73	1
 74	1
 75	0
-76	0
-77	0
+76	1
+77	1
 78	0
 79	1
 80	1
@@ -107,7 +107,7 @@ index	prediction
 105	1
 106	0
 107	1
-108	1
+108	0
 109	1
 110	0
 111	0
@@ -133,7 +133,7 @@ index	prediction
 131	1
 132	0
 133	0
-134	0
+134	1
 135	1
 136	1
 137	1
@@ -148,7 +148,7 @@ index	prediction
 146	0
 147	1
 148	1
-149	1
+149	0
 150	0
 151	1
 152	0
@@ -162,11 +162,11 @@ index	prediction
 160	1
 161	0
 162	0
-163	1
+163	0
 164	1
-165	0
+165	1
 166	0
-167	1
+167	0
 168	0
 169	1
 170	1
@@ -187,17 +187,17 @@ index	prediction
 185	1
 186	0
 187	0
-188	0
+188	1
 189	0
 190	0
 191	0
-192	1
+192	0
 193	1
 194	0
 195	1
 196	0
 197	1
-198	1
+198	0
 199	1
 200	0
 201	1
@@ -218,7 +218,7 @@ index	prediction
 216	0
 217	0
 218	1
-219	0
+219	1
 220	1
 221	0
 222	1

runs/Jul05_08-58-57_0a840cd07d02/events.out.tfevents.1688548330.0a840cd07d02.804.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0932018f9cd7bb8d4c7eae2371dcae97e28e0102b2021870b1ea0cd5abd1ab0b
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
-    "train_loss": 0.18657105684280395,
-    "train_runtime": 93.1154,
+    "train_loss": 0.17726608804294042,
+    "train_runtime": 611.8323,
     "train_samples": 1120,
-    "train_samples_per_second": 60.14,
-    "train_steps_per_second": 3.759
+    "train_samples_per_second": 9.153,
+    "train_steps_per_second": 1.144
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.8908296943231441,
-  "best_model_checkpoint": "models/fake-news-bert-base-spanish-wwm-cased/checkpoint-210",
+  "best_metric": 0.8968609865470851,
+  "best_model_checkpoint": "models/fake-news-bert-base-spanish-wwm-cased/checkpoint-420",
   "epoch": 5.0,
-  "global_step": 350,
+  "global_step": 700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -10,89 +10,89 @@
     {
       "epoch": 1.0,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 0.512,
-      "step": 70
+      "loss": 0.5098,
+      "step": 140
     },
     {
       "epoch": 1.0,
-      "eval_f1": 0.7473684210526316,
-      "eval_loss": 0.5346518158912659,
-      "eval_runtime": 1.3022,
-      "eval_samples_per_second": 184.298,
-      "eval_steps_per_second": 11.519,
-      "step": 70
+      "eval_f1": 0.7340425531914895,
+      "eval_loss": 0.49776777625083923,
+      "eval_runtime": 8.4133,
+      "eval_samples_per_second": 28.526,
+      "eval_steps_per_second": 3.566,
+      "step": 140
     },
     {
       "epoch": 2.0,
       "learning_rate": 1.2e-05,
-      "loss": 0.2281,
-      "step": 140
+      "loss": 0.2473,
+      "step": 280
     },
     {
       "epoch": 2.0,
-      "eval_f1": 0.8333333333333334,
-      "eval_loss": 0.3819647431373596,
-      "eval_runtime": 1.3101,
-      "eval_samples_per_second": 183.197,
-      "eval_steps_per_second": 11.45,
-      "step": 140
+      "eval_f1": 0.8828828828828829,
+      "eval_loss": 0.38877564668655396,
+      "eval_runtime": 8.625,
+      "eval_samples_per_second": 27.826,
+      "eval_steps_per_second": 3.478,
+      "step": 280
     },
     {
       "epoch": 3.0,
       "learning_rate": 8.000000000000001e-06,
-      "loss": 0.1212,
-      "step": 210
+      "loss": 0.0908,
+      "step": 420
     },
     {
       "epoch": 3.0,
-      "eval_f1": 0.8908296943231441,
-      "eval_loss": 0.3823632299900055,
-      "eval_runtime": 1.3116,
-      "eval_samples_per_second": 182.984,
-      "eval_steps_per_second": 11.437,
-      "step": 210
+      "eval_f1": 0.8968609865470851,
+      "eval_loss": 0.44203802943229675,
+      "eval_runtime": 8.4917,
+      "eval_samples_per_second": 28.263,
+      "eval_steps_per_second": 3.533,
+      "step": 420
     },
     {
       "epoch": 4.0,
       "learning_rate": 4.000000000000001e-06,
-      "loss": 0.0495,
-      "step": 280
+      "loss": 0.0332,
+      "step": 560
     },
     {
       "epoch": 4.0,
-      "eval_f1": 0.8818181818181818,
-      "eval_loss": 0.4416821300983429,
-      "eval_runtime": 1.3226,
-      "eval_samples_per_second": 181.456,
-      "eval_steps_per_second": 11.341,
-      "step": 280
+      "eval_f1": 0.8796296296296298,
+      "eval_loss": 0.5603806376457214,
+      "eval_runtime": 8.5449,
+      "eval_samples_per_second": 28.087,
+      "eval_steps_per_second": 3.511,
+      "step": 560
     },
     {
       "epoch": 5.0,
       "learning_rate": 0.0,
-      "loss": 0.0221,
-      "step": 350
+      "loss": 0.0052,
+      "step": 700
     },
     {
       "epoch": 5.0,
-      "eval_f1": 0.8767123287671234,
-      "eval_loss": 0.504033088684082,
-      "eval_runtime": 1.3358,
-      "eval_samples_per_second": 179.671,
-      "eval_steps_per_second": 11.229,
-      "step": 350
+      "eval_f1": 0.8956521739130435,
+      "eval_loss": 0.5154075622558594,
+      "eval_runtime": 8.6086,
+      "eval_samples_per_second": 27.879,
+      "eval_steps_per_second": 3.485,
+      "step": 700
     },
     {
       "epoch": 5.0,
-      "step": 350,
+      "step": 700,
       "total_flos": 1473421910016000.0,
-      "train_loss": 0.18657105684280395,
-      "train_runtime": 93.1154,
-      "train_samples_per_second": 60.14,
-      "train_steps_per_second": 3.759
+      "train_loss": 0.17726608804294042,
+      "train_runtime": 611.8323,
+      "train_samples_per_second": 9.153,
+      "train_steps_per_second": 1.144
     }
   ],
-  "max_steps": 350,
+  "max_steps": 700,
   "num_train_epochs": 5,
   "total_flos": 1473421910016000.0,
   "trial_name": null,