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": 66.66666666666667,
-    "f1": 76.95180617275538,
-    "train_loss": 0.5657880417844082,
-    "train_runtime": 541.4754,
+    "exact_match": 68.63333333333334,
+    "f1": 78.6554749331903,
+    "train_loss": 0.3079620914256319,
+    "train_runtime": 1090.9343,
     "train_samples": 3000,
-    "train_samples_per_second": 55.404,
-    "train_steps_per_second": 0.868
+    "train_samples_per_second": 54.999,
+    "train_steps_per_second": 0.862
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:649bfecda12824a048914781ee05ba174f35d1b7b23b8a9b01df02a45076cce8
+oid sha256:a17321f49e703bb81da7d55c6b77f3c2a57d1b900084a68db93a1f13b7229315
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65615ed4fe1c8c9e41a22dc324bdb4e22c39c9726ed41e382bad39176689b6a3
-size 14067254
+oid sha256:0ea47c0e2e0ea40f8eca67d559c18ab76ad97bab6a0bc746a37cbb327b928865
+size 14152806

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 10.0,
+    "epoch": 20.0,
     "eval_samples": 3000,
-    "exact_match": 66.66666666666667,
-    "f1": 76.95180617275538
+    "exact_match": 68.63333333333334,
+    "f1": 78.6554749331903
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.5657880417844082,
-    "train_runtime": 541.4754,
+    "epoch": 20.0,
+    "train_loss": 0.3079620914256319,
+    "train_runtime": 1090.9343,
     "train_samples": 3000,
-    "train_samples_per_second": 55.404,
-    "train_steps_per_second": 0.868
+    "train_samples_per_second": 54.999,
+    "train_steps_per_second": 0.862
 }

trainer_state.json

@@ -1,28 +1,34 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
+  "epoch": 20.0,
   "eval_steps": 500,
-  "global_step": 470,
+  "global_step": 940,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 10.0,
-      "step": 470,
-      "total_flos": 2.089590732288e+16,
-      "train_loss": 0.5657880417844082,
-      "train_runtime": 541.4754,
-      "train_samples_per_second": 55.404,
-      "train_steps_per_second": 0.868
+      "epoch": 10.64,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.5637,
+      "step": 500
+    },
+    {
+      "epoch": 20.0,
+      "step": 940,
+      "total_flos": 4.179181464576e+16,
+      "train_loss": 0.3079620914256319,
+      "train_runtime": 1090.9343,
+      "train_samples_per_second": 54.999,
+      "train_steps_per_second": 0.862
     }
   ],
   "logging_steps": 500,
-  "max_steps": 470,
-  "num_train_epochs": 10,
+  "max_steps": 940,
+  "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 2.089590732288e+16,
+  "total_flos": 4.179181464576e+16,
   "trial_name": null,
   "trial_params": null
 }