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": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 68.3,
-    "f1": 77.35539073416115,
-    "train_loss": 0.1694981136669715,
-    "train_runtime": 2376.6749,
+    "exact_match": 68.86666666666666,
+    "f1": 77.76425661888325,
+    "train_loss": 0.09819502154986064,
+    "train_runtime": 4749.1078,
     "train_samples": 3000,
-    "train_samples_per_second": 63.113,
+    "train_samples_per_second": 63.17,
     "train_steps_per_second": 0.505
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bce3eea016a0c429101ebfc0f9b360212928d83b1f55f34d6e88d07ae2bcb92b
+oid sha256:77d11f566704b8ea970a8b9270c4a9beee63dee776334b5e5fc069c5088dbf0c
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6763bbcbd8b21d7e3fed94b03bacdb4881ee275f634f08a710c42512c94e46f3
-size 13811032
+oid sha256:0a5bfba2d642d62d6d3840b324816dc1169c1ca3f0e889b0b8dc530396b98917
+size 13896163

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 68.3,
-    "f1": 77.35539073416115
+    "exact_match": 68.86666666666666,
+    "f1": 77.76425661888325
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.1694981136669715,
-    "train_runtime": 2376.6749,
+    "epoch": 100.0,
+    "train_loss": 0.09819502154986064,
+    "train_runtime": 4749.1078,
     "train_samples": 3000,
-    "train_samples_per_second": 63.113,
+    "train_samples_per_second": 63.17,
     "train_steps_per_second": 0.505
 }

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": 0.3975,
+      "learning_rate": 3.958333333333333e-05,
+      "loss": 0.4465,
       "step": 500
     },
     {
       "epoch": 41.67,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.0083,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.0159,
       "step": 1000
     },
     {
-      "epoch": 50.0,
-      "step": 1200,
-      "total_flos": 1.044795366144e+17,
-      "train_loss": 0.1694981136669715,
-      "train_runtime": 2376.6749,
-      "train_samples_per_second": 63.113,
+      "epoch": 62.5,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.005,
+      "step": 1500
+    },
+    {
+      "epoch": 83.33,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0025,
+      "step": 2000
+    },
+    {
+      "epoch": 100.0,
+      "step": 2400,
+      "total_flos": 2.089590732288e+17,
+      "train_loss": 0.09819502154986064,
+      "train_runtime": 4749.1078,
+      "train_samples_per_second": 63.17,
       "train_steps_per_second": 0.505
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1200,
-  "num_train_epochs": 50,
+  "max_steps": 2400,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 1.044795366144e+17,
+  "total_flos": 2.089590732288e+17,
   "trial_name": null,
   "trial_params": null
 }