End of training

README.md

@@ -2,7 +2,10 @@
 license: apache-2.0
 base_model: bert-large-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-large-uncased](https://huggingface.co/bert-large-uncased) on the None dataset.
+This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the squad dataset.
 
 ## Model description
 

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 10.0,
+    "epoch": 20.0,
     "eval_samples": 3000,
-    "exact_match": 68.73333333333333,
-    "f1": 78.36309301342095,
-    "train_loss": 0.48002195561185795,
-    "train_runtime": 672.6088,
+    "exact_match": 69.03333333333333,
+    "f1": 78.21299403265996,
+    "train_loss": 0.2561215139449911,
+    "train_runtime": 1328.0047,
     "train_samples": 3000,
-    "train_samples_per_second": 44.602,
-    "train_steps_per_second": 1.398
+    "train_samples_per_second": 45.181,
+    "train_steps_per_second": 1.416
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4dd6dd4ca7a845bc22a9d743c4543b7f3f59c7a9c7ea6b4267216a8e8fa7fad
+oid sha256:ef08c2a2a968ce0542edfbe8738b35698407b9918120a509c013b89ffc2b08f9
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f523387d37ce6d7a45723c9adc184a901b07da45282cb4e461a24a8c77645f3
-size 13889786
+oid sha256:ca84f218dbedfc47fa3f7f6ac34b6f85ff675325944c60c9c8fcb02eda633c74
+size 13785669

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 10.0,
+    "epoch": 20.0,
     "eval_samples": 3000,
-    "exact_match": 68.73333333333333,
-    "f1": 78.36309301342095
+    "exact_match": 69.03333333333333,
+    "f1": 78.21299403265996
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.48002195561185795,
-    "train_runtime": 672.6088,
+    "epoch": 20.0,
+    "train_loss": 0.2561215139449911,
+    "train_runtime": 1328.0047,
     "train_samples": 3000,
-    "train_samples_per_second": 44.602,
-    "train_steps_per_second": 1.398
+    "train_samples_per_second": 45.181,
+    "train_steps_per_second": 1.416
 }

trainer_state.json

@@ -1,34 +1,46 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
+  "epoch": 20.0,
   "eval_steps": 500,
-  "global_step": 940,
+  "global_step": 1880,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 5.32,
-      "learning_rate": 2.340425531914894e-05,
-      "loss": 0.8504,
+      "learning_rate": 3.670212765957447e-05,
+      "loss": 0.8422,
       "step": 500
     },
     {
-      "epoch": 10.0,
-      "step": 940,
-      "total_flos": 2.089590732288e+16,
-      "train_loss": 0.48002195561185795,
-      "train_runtime": 672.6088,
-      "train_samples_per_second": 44.602,
-      "train_steps_per_second": 1.398
+      "epoch": 10.64,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.0931,
+      "step": 1000
+    },
+    {
+      "epoch": 15.96,
+      "learning_rate": 1.0106382978723404e-05,
+      "loss": 0.0236,
+      "step": 1500
+    },
+    {
+      "epoch": 20.0,
+      "step": 1880,
+      "total_flos": 4.179181464576e+16,
+      "train_loss": 0.2561215139449911,
+      "train_runtime": 1328.0047,
+      "train_samples_per_second": 45.181,
+      "train_steps_per_second": 1.416
     }
   ],
   "logging_steps": 500,
-  "max_steps": 940,
-  "num_train_epochs": 10,
+  "max_steps": 1880,
+  "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 2.089590732288e+16,
+  "total_flos": 4.179181464576e+16,
   "trial_name": null,
   "trial_params": null
 }