End of training

README.md

@@ -2,7 +2,10 @@
 license: apache-2.0
 base_model: bert-large-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-large-uncased](https://huggingface.co/bert-large-uncased) on the None dataset.
+This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the squad dataset.
 
 ## Model description
 

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 100.0,
+    "epoch": 3.0,
     "eval_samples": 3000,
-    "exact_match": 61.7,
-    "f1": 72.37641409390515,
-    "train_loss": 0.08505555467402681,
-    "train_runtime": 5448.3986,
+    "exact_match": 65.33333333333333,
+    "f1": 75.53040749041759,
+    "train_loss": 1.3561742255028257,
+    "train_runtime": 201.7662,
     "train_samples": 3000,
-    "train_samples_per_second": 55.062,
-    "train_steps_per_second": 0.863
+    "train_samples_per_second": 44.606,
+    "train_steps_per_second": 1.398
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e784de55c8b6496e0ddaf8be0b4efb99adb377403f35b225697a7238e71a5946
+oid sha256:426dde7dbdce52a9f39b948f32973d344f00884e5c0e3203c9f7fe90b14746d2
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90b85b6d256166f8c675e5063c13bef00a34c9aafa8c7d17bda2b2179ab62680
-size 13888787
+oid sha256:c1e06574c62ad947e8cf9ed2283dae9b0e7609d0692b817f3635106e5001c47a
+size 13778063

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 100.0,
+    "epoch": 3.0,
     "eval_samples": 3000,
-    "exact_match": 61.7,
-    "f1": 72.37641409390515
+    "exact_match": 65.33333333333333,
+    "f1": 75.53040749041759
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 100.0,
-    "train_loss": 0.08505555467402681,
-    "train_runtime": 5448.3986,
+    "epoch": 3.0,
+    "train_loss": 1.3561742255028257,
+    "train_runtime": 201.7662,
     "train_samples": 3000,
-    "train_samples_per_second": 55.062,
-    "train_steps_per_second": 0.863
+    "train_samples_per_second": 44.606,
+    "train_steps_per_second": 1.398
 }

trainer_state.json

@@ -1,82 +1,28 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 100.0,
+  "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 4700,
+  "global_step": 282,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 10.64,
-      "learning_rate": 4.468085106382979e-05,
-      "loss": 0.634,
-      "step": 500
-    },
-    {
-      "epoch": 21.28,
-      "learning_rate": 3.936170212765958e-05,
-      "loss": 0.0578,
-      "step": 1000
-    },
-    {
-      "epoch": 31.91,
-      "learning_rate": 3.4042553191489365e-05,
-      "loss": 0.0315,
-      "step": 1500
-    },
-    {
-      "epoch": 42.55,
-      "learning_rate": 2.8723404255319154e-05,
-      "loss": 0.0148,
-      "step": 2000
-    },
-    {
-      "epoch": 53.19,
-      "learning_rate": 2.340425531914894e-05,
-      "loss": 0.0092,
-      "step": 2500
-    },
-    {
-      "epoch": 63.83,
-      "learning_rate": 1.8085106382978724e-05,
-      "loss": 0.0083,
-      "step": 3000
-    },
-    {
-      "epoch": 74.47,
-      "learning_rate": 1.2765957446808511e-05,
-      "loss": 0.0094,
-      "step": 3500
-    },
-    {
-      "epoch": 85.11,
-      "learning_rate": 7.446808510638298e-06,
-      "loss": 0.0082,
-      "step": 4000
-    },
-    {
-      "epoch": 95.74,
-      "learning_rate": 2.1276595744680853e-06,
-      "loss": 0.0146,
-      "step": 4500
-    },
-    {
-      "epoch": 100.0,
-      "step": 4700,
-      "total_flos": 2.089590732288e+17,
-      "train_loss": 0.08505555467402681,
-      "train_runtime": 5448.3986,
-      "train_samples_per_second": 55.062,
-      "train_steps_per_second": 0.863
+      "epoch": 3.0,
+      "step": 282,
+      "total_flos": 6268772196864000.0,
+      "train_loss": 1.3561742255028257,
+      "train_runtime": 201.7662,
+      "train_samples_per_second": 44.606,
+      "train_steps_per_second": 1.398
     }
   ],
   "logging_steps": 500,
-  "max_steps": 4700,
-  "num_train_epochs": 100,
+  "max_steps": 282,
+  "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 2.089590732288e+17,
+  "total_flos": 6268772196864000.0,
   "trial_name": null,
   "trial_params": null
 }