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": 67.73333333333333,
-    "f1": 77.2388500051414,
-    "train_loss": 0.1448818277044499,
-    "train_runtime": 2740.5707,
+    "exact_match": 61.03333333333333,
+    "f1": 70.85304837358167,
+    "train_loss": 0.08927536832525375,
+    "train_runtime": 5468.8371,
     "train_samples": 3000,
-    "train_samples_per_second": 54.733,
-    "train_steps_per_second": 0.857
+    "train_samples_per_second": 54.856,
+    "train_steps_per_second": 0.859
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6ded28041678f1221250e1a2ab9808e7116dc0f5d5a7c32bd8c24ecea4cf0b6a
+oid sha256:429324931628e012a527a71de7be8c84b185e43fb715726153522cfd44fb6459
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7bc0c608b17a6a839affd6c62ce9c8bd780b446d07d2aa3bd8829f2cb2807a8
-size 13876260
+oid sha256:696ec9ff64aa083c3131dd4ea56447e268234e99b5c520ddd1fdaf3df073997c
+size 13881990

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 67.73333333333333,
-    "f1": 77.2388500051414
+    "exact_match": 61.03333333333333,
+    "f1": 70.85304837358167
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.1448818277044499,
-    "train_runtime": 2740.5707,
+    "epoch": 100.0,
+    "train_loss": 0.08927536832525375,
+    "train_runtime": 5468.8371,
     "train_samples": 3000,
-    "train_samples_per_second": 54.733,
-    "train_steps_per_second": 0.857
+    "train_samples_per_second": 54.856,
+    "train_steps_per_second": 0.859
 }

trainer_state.json

@@ -1,52 +1,82 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 2350,
+  "global_step": 4700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 10.64,
-      "learning_rate": 3.936170212765958e-05,
-      "loss": 0.6031,
+      "learning_rate": 4.468085106382979e-05,
+      "loss": 0.634,
       "step": 500
     },
     {
       "epoch": 21.28,
-      "learning_rate": 2.8723404255319154e-05,
-      "loss": 0.0568,
+      "learning_rate": 3.936170212765958e-05,
+      "loss": 0.0578,
       "step": 1000
     },
     {
       "epoch": 31.91,
-      "learning_rate": 1.8085106382978724e-05,
-      "loss": 0.0144,
+      "learning_rate": 3.4042553191489365e-05,
+      "loss": 0.0315,
       "step": 1500
     },
     {
       "epoch": 42.55,
-      "learning_rate": 7.446808510638298e-06,
-      "loss": 0.0046,
+      "learning_rate": 2.8723404255319154e-05,
+      "loss": 0.0173,
       "step": 2000
     },
     {
-      "epoch": 50.0,
-      "step": 2350,
-      "total_flos": 1.044795366144e+17,
-      "train_loss": 0.1448818277044499,
-      "train_runtime": 2740.5707,
-      "train_samples_per_second": 54.733,
-      "train_steps_per_second": 0.857
+      "epoch": 53.19,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.0102,
+      "step": 2500
+    },
+    {
+      "epoch": 63.83,
+      "learning_rate": 1.8085106382978724e-05,
+      "loss": 0.0112,
+      "step": 3000
+    },
+    {
+      "epoch": 74.47,
+      "learning_rate": 1.2765957446808511e-05,
+      "loss": 0.0096,
+      "step": 3500
+    },
+    {
+      "epoch": 85.11,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 0.0131,
+      "step": 4000
+    },
+    {
+      "epoch": 95.74,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 0.0227,
+      "step": 4500
+    },
+    {
+      "epoch": 100.0,
+      "step": 4700,
+      "total_flos": 2.089590732288e+17,
+      "train_loss": 0.08927536832525375,
+      "train_runtime": 5468.8371,
+      "train_samples_per_second": 54.856,
+      "train_steps_per_second": 0.859
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2350,
-  "num_train_epochs": 50,
+  "max_steps": 4700,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 1.044795366144e+17,
+  "total_flos": 2.089590732288e+17,
   "trial_name": null,
   "trial_params": null
 }