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": 64.96666666666667,
-    "f1": 75.131491982848,
-    "train_loss": 0.8044876098632813,
-    "train_runtime": 901.6911,
+    "exact_match": 67.2,
+    "f1": 77.09945992044638,
+    "train_loss": 0.40876169204711915,
+    "train_runtime": 1551.6664,
     "train_samples": 3000,
-    "train_samples_per_second": 33.271,
-    "train_steps_per_second": 0.266
+    "train_samples_per_second": 38.668,
+    "train_steps_per_second": 0.309
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d27ae559d5d260af931f45075eacdbfa1e5d87690c1370bc5ddbe6f29eefdf75
+oid sha256:69c2dd007aa7a4e8db0443689c045404728a3643795e65cd01a5ce55de885c36
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:046f441c182f373db70a68e4485eed2bd8ddea921270f078e56218c113c60893
-size 13870028
+oid sha256:0149ec0ca7bfeee41220a704e7e726b3648029a0c4fdd1dbc13d44e48846b7f0
+size 13780699

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 10.0,
+    "epoch": 20.0,
     "eval_samples": 3000,
-    "exact_match": 64.96666666666667,
-    "f1": 75.131491982848
+    "exact_match": 67.2,
+    "f1": 77.09945992044638
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.8044876098632813,
-    "train_runtime": 901.6911,
+    "epoch": 20.0,
+    "train_loss": 0.40876169204711915,
+    "train_runtime": 1551.6664,
     "train_samples": 3000,
-    "train_samples_per_second": 33.271,
-    "train_steps_per_second": 0.266
+    "train_samples_per_second": 38.668,
+    "train_steps_per_second": 0.309
 }

trainer_state.json

@@ -1,28 +1,28 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
+  "epoch": 20.0,
   "eval_steps": 500,
-  "global_step": 240,
+  "global_step": 480,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 10.0,
-      "step": 240,
-      "total_flos": 2.089590732288e+16,
-      "train_loss": 0.8044876098632813,
-      "train_runtime": 901.6911,
-      "train_samples_per_second": 33.271,
-      "train_steps_per_second": 0.266
+      "epoch": 20.0,
+      "step": 480,
+      "total_flos": 4.179181464576e+16,
+      "train_loss": 0.40876169204711915,
+      "train_runtime": 1551.6664,
+      "train_samples_per_second": 38.668,
+      "train_steps_per_second": 0.309
     }
   ],
   "logging_steps": 500,
-  "max_steps": 240,
-  "num_train_epochs": 10,
+  "max_steps": 480,
+  "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 2.089590732288e+16,
+  "total_flos": 4.179181464576e+16,
   "trial_name": null,
   "trial_params": null
 }