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": 50.0,
+    "epoch": 100.0,
     "eval_samples": 5000,
-    "exact_match": 62.22,
-    "f1": 71.95882328661979,
-    "train_loss": 0.25276324987411497,
-    "train_runtime": 1413.9494,
+    "exact_match": 62.42,
+    "f1": 71.92532975760962,
+    "train_loss": 0.12865829187631608,
+    "train_runtime": 2954.7897,
     "train_samples": 5000,
-    "train_samples_per_second": 176.81,
-    "train_steps_per_second": 0.707
+    "train_samples_per_second": 169.217,
+    "train_steps_per_second": 0.677
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02786a2b8aa31481648ebb5d7d75d824eacfc2bc64318620127cdd232f9dc33b
+oid sha256:60778c7184b0f0b5a4081d060f40744a0f5106a72172afd3a2cebf89a4400b72
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9dd2c89529276e53a056a5cf5c51956d5c29915008e2c2bd400e554208e678b7
-size 23209136
+oid sha256:4ae5a7308bd4d718b30802f9273823a2c8ae475eba29fa1c7a77b547a62d06a1
+size 23139087

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 5000,
-    "exact_match": 62.22,
-    "f1": 71.95882328661979
+    "exact_match": 62.42,
+    "f1": 71.92532975760962
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.25276324987411497,
-    "train_runtime": 1413.9494,
+    "epoch": 100.0,
+    "train_loss": 0.12865829187631608,
+    "train_runtime": 2954.7897,
     "train_samples": 5000,
-    "train_samples_per_second": 176.81,
-    "train_steps_per_second": 0.707
+    "train_samples_per_second": 169.217,
+    "train_steps_per_second": 0.677
 }

trainer_state.json

@@ -1,40 +1,52 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 1000,
+  "global_step": 2000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 25.0,
-      "learning_rate": 2.5e-05,
-      "loss": 0.4919,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.4884,
       "step": 500
     },
     {
       "epoch": 50.0,
-      "learning_rate": 0.0,
-      "loss": 0.0136,
+      "learning_rate": 2.5e-05,
+      "loss": 0.0171,
       "step": 1000
     },
     {
-      "epoch": 50.0,
-      "step": 1000,
-      "total_flos": 4.8993141888e+16,
-      "train_loss": 0.25276324987411497,
-      "train_runtime": 1413.9494,
-      "train_samples_per_second": 176.81,
-      "train_steps_per_second": 0.707
+      "epoch": 75.0,
+      "learning_rate": 1.25e-05,
+      "loss": 0.0058,
+      "step": 1500
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 0.0,
+      "loss": 0.0033,
+      "step": 2000
+    },
+    {
+      "epoch": 100.0,
+      "step": 2000,
+      "total_flos": 9.7986283776e+16,
+      "train_loss": 0.12865829187631608,
+      "train_runtime": 2954.7897,
+      "train_samples_per_second": 169.217,
+      "train_steps_per_second": 0.677
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1000,
-  "num_train_epochs": 50,
+  "max_steps": 2000,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 4.8993141888e+16,
+  "total_flos": 9.7986283776e+16,
   "trial_name": null,
   "trial_params": null
 }