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": 10.0,
+    "epoch": 20.0,
     "eval_samples": 3000,
-    "exact_match": 60.2,
-    "f1": 70.56217910315665,
-    "train_loss": 0.6274924014477019,
-    "train_runtime": 264.4927,
+    "exact_match": 61.43333333333333,
+    "f1": 71.51754593320108,
+    "train_loss": 0.3228847931040094,
+    "train_runtime": 521.4493,
     "train_samples": 3000,
-    "train_samples_per_second": 113.425,
-    "train_steps_per_second": 3.554
+    "train_samples_per_second": 115.064,
+    "train_steps_per_second": 3.605
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da503632315f0270126891513a7212481030886142e9ad14823c0e5c1dbfd58a
+oid sha256:0d793897c3f3e821725f81e31136d37cba40e472f1b1ca00f99a6b764ba728b5
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e7b0af6c461862f9cef77ac33fe3cfcd3bdf5ba7424558b09e2b39755aad680
-size 14082921
+oid sha256:95601525f12ee20569c9a55526096f75e030a73f2b8cf984f4ffe9a598cd62e3
+size 14129938

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 10.0,
+    "epoch": 20.0,
     "eval_samples": 3000,
-    "exact_match": 60.2,
-    "f1": 70.56217910315665
+    "exact_match": 61.43333333333333,
+    "f1": 71.51754593320108
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.6274924014477019,
-    "train_runtime": 264.4927,
+    "epoch": 20.0,
+    "train_loss": 0.3228847931040094,
+    "train_runtime": 521.4493,
     "train_samples": 3000,
-    "train_samples_per_second": 113.425,
-    "train_steps_per_second": 3.554
+    "train_samples_per_second": 115.064,
+    "train_steps_per_second": 3.605
 }

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": 1.1006,
+      "learning_rate": 3.670212765957447e-05,
+      "loss": 1.0779,
       "step": 500
     },
     {
-      "epoch": 10.0,
-      "step": 940,
-      "total_flos": 5879177026560000.0,
-      "train_loss": 0.6274924014477019,
-      "train_runtime": 264.4927,
-      "train_samples_per_second": 113.425,
-      "train_steps_per_second": 3.554
+      "epoch": 10.64,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.0968,
+      "step": 1000
+    },
+    {
+      "epoch": 15.96,
+      "learning_rate": 1.0106382978723404e-05,
+      "loss": 0.0322,
+      "step": 1500
+    },
+    {
+      "epoch": 20.0,
+      "step": 1880,
+      "total_flos": 1.175835405312e+16,
+      "train_loss": 0.3228847931040094,
+      "train_runtime": 521.4493,
+      "train_samples_per_second": 115.064,
+      "train_steps_per_second": 3.605
     }
   ],
   "logging_steps": 500,
-  "max_steps": 940,
-  "num_train_epochs": 10,
+  "max_steps": 1880,
+  "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 5879177026560000.0,
+  "total_flos": 1.175835405312e+16,
   "trial_name": null,
   "trial_params": null
 }