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": 59.96666666666667,
-    "f1": 70.27838709335163,
-    "train_loss": 0.24115766083200774,
-    "train_runtime": 850.2688,
+    "exact_match": 60.8,
+    "f1": 70.30919492004044,
+    "train_loss": 0.12285581573843957,
+    "train_runtime": 1701.0115,
     "train_samples": 3000,
-    "train_samples_per_second": 176.415,
+    "train_samples_per_second": 176.366,
     "train_steps_per_second": 1.411
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c6a99b4a1591942a76904055e7068b6071213c5832230534286a936028dc2dc
+oid sha256:b06caf0a142814dcd5dcee1eb93268b11ac3fb79364bf924cc3e2d9010eb7497
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc4cb0cb213e161395b45bc501174d0a5e88d7769d39e18b40747bec8e498602
-size 14225017
+oid sha256:915ea74c899c8df2a430313a8320707f212956d40e466edc8ae7c9300dcd46a6
+size 14125601

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 59.96666666666667,
-    "f1": 70.27838709335163
+    "exact_match": 60.8,
+    "f1": 70.30919492004044
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.24115766083200774,
-    "train_runtime": 850.2688,
+    "epoch": 100.0,
+    "train_loss": 0.12285581573843957,
+    "train_runtime": 1701.0115,
     "train_samples": 3000,
-    "train_samples_per_second": 176.415,
+    "train_samples_per_second": 176.366,
     "train_steps_per_second": 1.411
 }

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": 1200,
+  "global_step": 2400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 20.83,
-      "learning_rate": 2.916666666666667e-05,
-      "loss": 0.5624,
+      "learning_rate": 3.958333333333333e-05,
+      "loss": 0.5594,
       "step": 500
     },
     {
       "epoch": 41.67,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.0146,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.0197,
       "step": 1000
     },
     {
-      "epoch": 50.0,
-      "step": 1200,
-      "total_flos": 2.93958851328e+16,
-      "train_loss": 0.24115766083200774,
-      "train_runtime": 850.2688,
-      "train_samples_per_second": 176.415,
+      "epoch": 62.5,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.0056,
+      "step": 1500
+    },
+    {
+      "epoch": 83.33,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0028,
+      "step": 2000
+    },
+    {
+      "epoch": 100.0,
+      "step": 2400,
+      "total_flos": 5.87917702656e+16,
+      "train_loss": 0.12285581573843957,
+      "train_runtime": 1701.0115,
+      "train_samples_per_second": 176.366,
       "train_steps_per_second": 1.411
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1200,
-  "num_train_epochs": 50,
+  "max_steps": 2400,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 2.93958851328e+16,
+  "total_flos": 5.87917702656e+16,
   "trial_name": null,
   "trial_params": null
 }