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": 50.0,
     "eval_samples": 3000,
-    "exact_match": 60.2,
-    "f1": 70.56217910315665,
-    "train_loss": 0.6274924014477019,
-    "train_runtime": 267.6611,
+    "exact_match": 55.833333333333336,
+    "f1": 65.67795577455873,
+    "train_loss": 0.1556171702324076,
+    "train_runtime": 1356.8935,
     "train_samples": 3000,
-    "train_samples_per_second": 112.082,
-    "train_steps_per_second": 3.512
+    "train_samples_per_second": 110.547,
+    "train_steps_per_second": 3.464
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:30dc4bfdecc4f8bda46f12340afa0f3283ef9640136a6ed75059db7badcb250f
+oid sha256:a6483b4338266a431a726740f9d0603e8af5b5ea355462122d055eb126bb9e1e
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e7b0af6c461862f9cef77ac33fe3cfcd3bdf5ba7424558b09e2b39755aad680
-size 14082921
+oid sha256:630b581cec250940167a383435d09347c398cba0c12a4e660ef2760f5d1b9877
+size 14106174

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 10.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 60.2,
-    "f1": 70.56217910315665
+    "exact_match": 55.833333333333336,
+    "f1": 65.67795577455873
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.6274924014477019,
-    "train_runtime": 267.6611,
+    "epoch": 50.0,
+    "train_loss": 0.1556171702324076,
+    "train_runtime": 1356.8935,
     "train_samples": 3000,
-    "train_samples_per_second": 112.082,
-    "train_steps_per_second": 3.512
+    "train_samples_per_second": 110.547,
+    "train_steps_per_second": 3.464
 }

trainer_state.json

@@ -1,34 +1,82 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
+  "epoch": 50.0,
   "eval_steps": 500,
-  "global_step": 940,
+  "global_step": 4700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 5.32,
-      "learning_rate": 2.340425531914894e-05,
-      "loss": 1.1006,
+      "learning_rate": 4.468085106382979e-05,
+      "loss": 1.1017,
       "step": 500
     },
     {
-      "epoch": 10.0,
-      "step": 940,
-      "total_flos": 5879177026560000.0,
-      "train_loss": 0.6274924014477019,
-      "train_runtime": 267.6611,
-      "train_samples_per_second": 112.082,
-      "train_steps_per_second": 3.512
+      "epoch": 10.64,
+      "learning_rate": 3.936170212765958e-05,
+      "loss": 0.1281,
+      "step": 1000
+    },
+    {
+      "epoch": 15.96,
+      "learning_rate": 3.4042553191489365e-05,
+      "loss": 0.0555,
+      "step": 1500
+    },
+    {
+      "epoch": 21.28,
+      "learning_rate": 2.8723404255319154e-05,
+      "loss": 0.027,
+      "step": 2000
+    },
+    {
+      "epoch": 26.6,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.0195,
+      "step": 2500
+    },
+    {
+      "epoch": 31.91,
+      "learning_rate": 1.8085106382978724e-05,
+      "loss": 0.0135,
+      "step": 3000
+    },
+    {
+      "epoch": 37.23,
+      "learning_rate": 1.2765957446808511e-05,
+      "loss": 0.0103,
+      "step": 3500
+    },
+    {
+      "epoch": 42.55,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 0.0173,
+      "step": 4000
+    },
+    {
+      "epoch": 47.87,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 0.0432,
+      "step": 4500
+    },
+    {
+      "epoch": 50.0,
+      "step": 4700,
+      "total_flos": 2.93958851328e+16,
+      "train_loss": 0.1556171702324076,
+      "train_runtime": 1356.8935,
+      "train_samples_per_second": 110.547,
+      "train_steps_per_second": 3.464
     }
   ],
   "logging_steps": 500,
-  "max_steps": 940,
-  "num_train_epochs": 10,
+  "max_steps": 4700,
+  "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 5879177026560000.0,
+  "total_flos": 2.93958851328e+16,
   "trial_name": null,
   "trial_params": null
 }