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": 5000,
-    "exact_match": 63.44,
-    "f1": 73.23936492780788,
-    "train_loss": 0.4993326762679276,
-    "train_runtime": 435.4109,
+    "exact_match": 61.26,
+    "f1": 71.13079061487772,
+    "train_loss": 0.27299969211505476,
+    "train_runtime": 864.6725,
     "train_samples": 5000,
-    "train_samples_per_second": 114.834,
-    "train_steps_per_second": 3.606
+    "train_samples_per_second": 115.651,
+    "train_steps_per_second": 3.631
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:582307e537606fc7e95a67ea720b3d66dd8e91c37fbb1573b9a4ef95e35bbb25
+oid sha256:94f861260e6b0d3f7ce7a2ae9f36c76dd6e0a042757d9608f2786a66d1f92a6e
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86fc5bb73d9f87545c3a6e4966872b53b6774b98e346fb7ef856261fc86ba95b
-size 23075256
+oid sha256:dcf358651efc785798b4452167185f1aebaee14c685697f64dd54d412814f7e8
+size 23190722

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 10.0,
+    "epoch": 20.0,
     "eval_samples": 5000,
-    "exact_match": 63.44,
-    "f1": 73.23936492780788
+    "exact_match": 61.26,
+    "f1": 71.13079061487772
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.4993326762679276,
-    "train_runtime": 435.4109,
+    "epoch": 20.0,
+    "train_loss": 0.27299969211505476,
+    "train_runtime": 864.6725,
     "train_samples": 5000,
-    "train_samples_per_second": 114.834,
-    "train_steps_per_second": 3.606
+    "train_samples_per_second": 115.651,
+    "train_steps_per_second": 3.631
 }

trainer_state.json

@@ -1,46 +1,64 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
+  "epoch": 20.0,
   "eval_steps": 500,
-  "global_step": 1570,
+  "global_step": 3140,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 3.18,
-      "learning_rate": 3.407643312101911e-05,
-      "loss": 1.2801,
+      "learning_rate": 4.2038216560509556e-05,
+      "loss": 1.2813,
       "step": 500
     },
     {
       "epoch": 6.37,
-      "learning_rate": 1.8152866242038215e-05,
-      "loss": 0.2111,
+      "learning_rate": 3.407643312101911e-05,
+      "loss": 0.2307,
       "step": 1000
     },
     {
       "epoch": 9.55,
-      "learning_rate": 2.229299363057325e-06,
-      "loss": 0.0713,
+      "learning_rate": 2.6114649681528662e-05,
+      "loss": 0.1029,
       "step": 1500
     },
     {
-      "epoch": 10.0,
-      "step": 1570,
-      "total_flos": 9798628377600000.0,
-      "train_loss": 0.4993326762679276,
-      "train_runtime": 435.4109,
-      "train_samples_per_second": 114.834,
-      "train_steps_per_second": 3.606
+      "epoch": 12.74,
+      "learning_rate": 1.8152866242038215e-05,
+      "loss": 0.0472,
+      "step": 2000
+    },
+    {
+      "epoch": 15.92,
+      "learning_rate": 1.0191082802547772e-05,
+      "loss": 0.0256,
+      "step": 2500
+    },
+    {
+      "epoch": 19.11,
+      "learning_rate": 2.229299363057325e-06,
+      "loss": 0.0184,
+      "step": 3000
+    },
+    {
+      "epoch": 20.0,
+      "step": 3140,
+      "total_flos": 1.95972567552e+16,
+      "train_loss": 0.27299969211505476,
+      "train_runtime": 864.6725,
+      "train_samples_per_second": 115.651,
+      "train_steps_per_second": 3.631
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1570,
-  "num_train_epochs": 10,
+  "max_steps": 3140,
+  "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 9798628377600000.0,
+  "total_flos": 1.95972567552e+16,
   "trial_name": null,
   "trial_params": null
 }