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": 3000,
-    "exact_match": 57.9,
-    "f1": 68.27498216863845,
-    "train_loss": 0.35042489926020304,
-    "train_runtime": 779.0866,
+    "exact_match": 60.166666666666664,
+    "f1": 70.36438968174092,
+    "train_loss": 0.17385060826937357,
+    "train_runtime": 1558.7791,
     "train_samples": 3000,
-    "train_samples_per_second": 192.533,
+    "train_samples_per_second": 192.458,
     "train_steps_per_second": 0.77
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54aa920f20e321e78a1eb5ebd6323d143b3c06104edd06414f76ad18ec36aee9
+oid sha256:bdaae201cbbedd58bc78e2165dea4ee886d0b4e661a4984e8ff7124e2bd3db86
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02d5734fc8cfd7b952811b535ff2e561117313ca5cddef2276e655026e330ac4
-size 14252229
+oid sha256:89f203ec52562a73b5fd902f89fb79713e24d5e2e376f6e0a72f609b6050efd3
+size 14194475

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 57.9,
-    "f1": 68.27498216863845
+    "exact_match": 60.166666666666664,
+    "f1": 70.36438968174092
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.35042489926020304,
-    "train_runtime": 779.0866,
+    "epoch": 100.0,
+    "train_loss": 0.17385060826937357,
+    "train_runtime": 1558.7791,
     "train_samples": 3000,
-    "train_samples_per_second": 192.533,
+    "train_samples_per_second": 192.458,
     "train_steps_per_second": 0.77
 }

trainer_state.json

@@ -1,34 +1,40 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 600,
+  "global_step": 1200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 41.67,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.4187,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.4101,
       "step": 500
     },
     {
-      "epoch": 50.0,
-      "step": 600,
-      "total_flos": 2.93958851328e+16,
-      "train_loss": 0.35042489926020304,
-      "train_runtime": 779.0866,
-      "train_samples_per_second": 192.533,
+      "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": 1558.7791,
+      "train_samples_per_second": 192.458,
       "train_steps_per_second": 0.77
     }
   ],
   "logging_steps": 500,
-  "max_steps": 600,
-  "num_train_epochs": 50,
+  "max_steps": 1200,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 2.93958851328e+16,
+  "total_flos": 5.87917702656e+16,
   "trial_name": null,
   "trial_params": null
 }