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": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 55.733333333333334,
-    "f1": 65.89806625454503,
-    "train_loss": 0.9096226374308268,
-    "train_runtime": 311.688,
+    "exact_match": 57.9,
+    "f1": 68.27498216863845,
+    "train_loss": 0.35042489926020304,
+    "train_runtime": 777.9309,
     "train_samples": 3000,
-    "train_samples_per_second": 192.5,
-    "train_steps_per_second": 0.77
+    "train_samples_per_second": 192.819,
+    "train_steps_per_second": 0.771
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ac7d785df01636ca4e69bda311caad1af9a4eef4157e178dabcca1a4c5538e4
+oid sha256:4343a6fd04166127c1a94516aa332e957d6cdbec8c9a8ec5671970414ff361aa
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2031d47beab2b9bb81c6a02b151f810d2d6fb8e7a53b77cf903a95ca7f7c9126
-size 14202765
+oid sha256:02d5734fc8cfd7b952811b535ff2e561117313ca5cddef2276e655026e330ac4
+size 14252229

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 55.733333333333334,
-    "f1": 65.89806625454503
+    "exact_match": 57.9,
+    "f1": 68.27498216863845
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.9096226374308268,
-    "train_runtime": 311.688,
+    "epoch": 50.0,
+    "train_loss": 0.35042489926020304,
+    "train_runtime": 777.9309,
     "train_samples": 3000,
-    "train_samples_per_second": 192.5,
-    "train_steps_per_second": 0.77
+    "train_samples_per_second": 192.819,
+    "train_steps_per_second": 0.771
 }

trainer_state.json

@@ -1,28 +1,34 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 20.0,
+  "epoch": 50.0,
   "eval_steps": 500,
-  "global_step": 240,
+  "global_step": 600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 20.0,
-      "step": 240,
-      "total_flos": 1.175835405312e+16,
-      "train_loss": 0.9096226374308268,
-      "train_runtime": 311.688,
-      "train_samples_per_second": 192.5,
-      "train_steps_per_second": 0.77
+      "epoch": 41.67,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.4187,
+      "step": 500
+    },
+    {
+      "epoch": 50.0,
+      "step": 600,
+      "total_flos": 2.93958851328e+16,
+      "train_loss": 0.35042489926020304,
+      "train_runtime": 777.9309,
+      "train_samples_per_second": 192.819,
+      "train_steps_per_second": 0.771
     }
   ],
   "logging_steps": 500,
-  "max_steps": 240,
-  "num_train_epochs": 20,
+  "max_steps": 600,
+  "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 1.175835405312e+16,
+  "total_flos": 2.93958851328e+16,
   "trial_name": null,
   "trial_params": null
 }