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": 3.0,
+    "epoch": 5.0,
     "eval_samples": 3000,
-    "exact_match": 57.63333333333333,
-    "f1": 68.29950643332113,
-    "train_loss": 2.4093204074435763,
-    "train_runtime": 149.3769,
+    "exact_match": 62.7,
+    "f1": 73.16174273334184,
+    "train_loss": 1.5756070454915365,
+    "train_runtime": 240.5686,
     "train_samples": 3000,
-    "train_samples_per_second": 60.25,
-    "train_steps_per_second": 0.482
+    "train_samples_per_second": 62.352,
+    "train_steps_per_second": 0.499
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23ea8312223d9b952653f1341dd86666f831f2fb6f4e46765101a4741e25da20
+oid sha256:28a8def7ae2e42b0293a62ee292e2fb5e7b68a3b0bccdca0a198cd111e78c801
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e2e0a226f5b7c433b1c836cc893773837fa72058a9a1c9191b08c08563d860b3
-size 13838402
+oid sha256:c86da686bf21a775f4c49d02866662d7df17012d64b1af095abaf5d12c827099
+size 13710196

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 3.0,
+    "epoch": 5.0,
     "eval_samples": 3000,
-    "exact_match": 57.63333333333333,
-    "f1": 68.29950643332113
+    "exact_match": 62.7,
+    "f1": 73.16174273334184
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "train_loss": 2.4093204074435763,
-    "train_runtime": 149.3769,
+    "epoch": 5.0,
+    "train_loss": 1.5756070454915365,
+    "train_runtime": 240.5686,
     "train_samples": 3000,
-    "train_samples_per_second": 60.25,
-    "train_steps_per_second": 0.482
+    "train_samples_per_second": 62.352,
+    "train_steps_per_second": 0.499
 }

trainer_state.json

@@ -1,28 +1,28 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.0,
+  "epoch": 5.0,
   "eval_steps": 500,
-  "global_step": 72,
+  "global_step": 120,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 3.0,
-      "step": 72,
-      "total_flos": 6268772196864000.0,
-      "train_loss": 2.4093204074435763,
-      "train_runtime": 149.3769,
-      "train_samples_per_second": 60.25,
-      "train_steps_per_second": 0.482
+      "epoch": 5.0,
+      "step": 120,
+      "total_flos": 1.044795366144e+16,
+      "train_loss": 1.5756070454915365,
+      "train_runtime": 240.5686,
+      "train_samples_per_second": 62.352,
+      "train_steps_per_second": 0.499
     }
   ],
   "logging_steps": 500,
-  "max_steps": 72,
-  "num_train_epochs": 3,
+  "max_steps": 120,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 6268772196864000.0,
+  "total_flos": 1.044795366144e+16,
   "trial_name": null,
   "trial_params": null
 }