End of training

README.md

@@ -2,7 +2,10 @@
 license: apache-2.0
 base_model: bert-base-uncased
 tags:
+- question-answering
 - generated_from_trainer
+datasets:
+- squad
 model-index:
 - name: out
   results: []
@@ -13,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # out
 
-This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
+This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the squad dataset.
 
 ## Model description
 

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 100.0,
+    "epoch": 3.0,
     "eval_samples": 5000,
-    "exact_match": 36.42,
-    "f1": 48.8636602149283,
-    "train_loss": 0.33136190450644193,
-    "train_runtime": 3703.1966,
+    "exact_match": 61.92,
+    "f1": 71.73903036462997,
+    "train_loss": 1.3744957796327628,
+    "train_runtime": 129.2382,
     "train_samples": 5000,
-    "train_samples_per_second": 135.018,
-    "train_steps_per_second": 2.133
+    "train_samples_per_second": 116.065,
+    "train_steps_per_second": 3.644
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:229db5bbfa3a421be49cd4b4e955121f1b3519ae667ccdad0ade4b4072c41386
+oid sha256:e1e8894b461f145375020b1fd6488849975e81926b1f3d1cc25564d4a8811bcc
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b555d8926b26c5a11d23aeefca95432e3f64ee07de0be9b9e8a2b22b1c225284
-size 22988932
+oid sha256:6ea2f532462c760f6c27d21337b3cabd1ecdf0f4bd44aa85cf898e6b8ef4aa21
+size 22776039

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 100.0,
+    "epoch": 3.0,
     "eval_samples": 5000,
-    "exact_match": 36.42,
-    "f1": 48.8636602149283
+    "exact_match": 61.92,
+    "f1": 71.73903036462997
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 100.0,
-    "train_loss": 0.33136190450644193,
-    "train_runtime": 3703.1966,
+    "epoch": 3.0,
+    "train_loss": 1.3744957796327628,
+    "train_runtime": 129.2382,
     "train_samples": 5000,
-    "train_samples_per_second": 135.018,
-    "train_steps_per_second": 2.133
+    "train_samples_per_second": 116.065,
+    "train_steps_per_second": 3.644
 }

trainer_state.json

@@ -1,118 +1,28 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 100.0,
+  "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 7900,
+  "global_step": 471,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 6.33,
-      "learning_rate": 4.683544303797468e-05,
-      "loss": 0.8991,
-      "step": 500
-    },
-    {
-      "epoch": 12.66,
-      "learning_rate": 4.367088607594937e-05,
-      "loss": 0.1014,
-      "step": 1000
-    },
-    {
-      "epoch": 18.99,
-      "learning_rate": 4.050632911392405e-05,
-      "loss": 0.0476,
-      "step": 1500
-    },
-    {
-      "epoch": 25.32,
-      "learning_rate": 3.7341772151898736e-05,
-      "loss": 0.0275,
-      "step": 2000
-    },
-    {
-      "epoch": 31.65,
-      "learning_rate": 3.4177215189873416e-05,
-      "loss": 0.0243,
-      "step": 2500
-    },
-    {
-      "epoch": 37.97,
-      "learning_rate": 3.10126582278481e-05,
-      "loss": 0.0165,
-      "step": 3000
-    },
-    {
-      "epoch": 44.3,
-      "learning_rate": 2.7848101265822786e-05,
-      "loss": 0.0252,
-      "step": 3500
-    },
-    {
-      "epoch": 50.63,
-      "learning_rate": 2.468354430379747e-05,
-      "loss": 0.034,
-      "step": 4000
-    },
-    {
-      "epoch": 56.96,
-      "learning_rate": 2.1518987341772153e-05,
-      "loss": 0.0469,
-      "step": 4500
-    },
-    {
-      "epoch": 63.29,
-      "learning_rate": 1.8354430379746836e-05,
-      "loss": 0.091,
-      "step": 5000
-    },
-    {
-      "epoch": 69.62,
-      "learning_rate": 1.5189873417721521e-05,
-      "loss": 0.2735,
-      "step": 5500
-    },
-    {
-      "epoch": 75.95,
-      "learning_rate": 1.2025316455696203e-05,
-      "loss": 0.4183,
-      "step": 6000
-    },
-    {
-      "epoch": 82.28,
-      "learning_rate": 8.860759493670886e-06,
-      "loss": 0.5516,
-      "step": 6500
-    },
-    {
-      "epoch": 88.61,
-      "learning_rate": 5.69620253164557e-06,
-      "loss": 0.6946,
-      "step": 7000
-    },
-    {
-      "epoch": 94.94,
-      "learning_rate": 2.531645569620253e-06,
-      "loss": 1.0278,
-      "step": 7500
-    },
-    {
-      "epoch": 100.0,
-      "step": 7900,
-      "total_flos": 9.7986283776e+16,
-      "train_loss": 0.33136190450644193,
-      "train_runtime": 3703.1966,
-      "train_samples_per_second": 135.018,
-      "train_steps_per_second": 2.133
+      "epoch": 3.0,
+      "step": 471,
+      "total_flos": 2939588513280000.0,
+      "train_loss": 1.3744957796327628,
+      "train_runtime": 129.2382,
+      "train_samples_per_second": 116.065,
+      "train_steps_per_second": 3.644
     }
   ],
   "logging_steps": 500,
-  "max_steps": 7900,
-  "num_train_epochs": 100,
+  "max_steps": 471,
+  "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 9.7986283776e+16,
+  "total_flos": 2939588513280000.0,
   "trial_name": null,
   "trial_params": null
 }