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": 100.0,
+    "epoch": 3.0,
     "eval_samples": 3000,
-    "exact_match": 60.166666666666664,
-    "f1": 70.36438968174092,
-    "train_loss": 0.17385060826937357,
-    "train_runtime": 1559.986,
+    "exact_match": 38.666666666666664,
+    "f1": 49.2849064548386,
+    "train_loss": 3.040504879421658,
+    "train_runtime": 55.3961,
     "train_samples": 3000,
-    "train_samples_per_second": 192.309,
-    "train_steps_per_second": 0.769
+    "train_samples_per_second": 162.466,
+    "train_steps_per_second": 1.3
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e01e16eb47e2159dfbc99ac02917f3ad3e483c426c2eab26937716e65e8c8537
+oid sha256:d401f61194855059e0d35939f96c3f64a2e7b5a3582b148acd3304864d8179f6
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89f203ec52562a73b5fd902f89fb79713e24d5e2e376f6e0a72f609b6050efd3
-size 14194475
+oid sha256:88b1907315e1e02f240f1e335087cf08aa5b03a289fcd063d12e937d73f51618
+size 14004394

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 100.0,
+    "epoch": 3.0,
     "eval_samples": 3000,
-    "exact_match": 60.166666666666664,
-    "f1": 70.36438968174092
+    "exact_match": 38.666666666666664,
+    "f1": 49.2849064548386
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 100.0,
-    "train_loss": 0.17385060826937357,
-    "train_runtime": 1559.986,
+    "epoch": 3.0,
+    "train_loss": 3.040504879421658,
+    "train_runtime": 55.3961,
     "train_samples": 3000,
-    "train_samples_per_second": 192.309,
-    "train_steps_per_second": 0.769
+    "train_samples_per_second": 162.466,
+    "train_steps_per_second": 1.3
 }

trainer_state.json

@@ -1,40 +1,28 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 100.0,
+  "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 1200,
+  "global_step": 72,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 41.67,
-      "learning_rate": 2.916666666666667e-05,
-      "loss": 0.4101,
-      "step": 500
-    },
-    {
-      "epoch": 83.33,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.0059,
-      "step": 1000
-    },
-    {
-      "epoch": 100.0,
-      "step": 1200,
-      "total_flos": 5.87917702656e+16,
-      "train_loss": 0.17385060826937357,
-      "train_runtime": 1559.986,
-      "train_samples_per_second": 192.309,
-      "train_steps_per_second": 0.769
+      "epoch": 3.0,
+      "step": 72,
+      "total_flos": 1763753107968000.0,
+      "train_loss": 3.040504879421658,
+      "train_runtime": 55.3961,
+      "train_samples_per_second": 162.466,
+      "train_steps_per_second": 1.3
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1200,
-  "num_train_epochs": 100,
+  "max_steps": 72,
+  "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 5.87917702656e+16,
+  "total_flos": 1763753107968000.0,
   "trial_name": null,
   "trial_params": null
 }