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": 20.0,
+    "epoch": 50.0,
     "eval_samples": 5000,
-    "exact_match": 62.5,
-    "f1": 72.40162540347497,
-    "train_loss": 0.32572943477690974,
-    "train_runtime": 684.9677,
+    "exact_match": 58.78,
+    "f1": 68.98441947756866,
+    "train_loss": 0.14661559768869908,
+    "train_runtime": 1698.5921,
     "train_samples": 5000,
-    "train_samples_per_second": 145.992,
-    "train_steps_per_second": 2.307
+    "train_samples_per_second": 147.181,
+    "train_steps_per_second": 2.325
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:900e2083858ea93ca01b25c2605bec4c3f470af6f5f84efc39ecba24591578c7
+oid sha256:8bf18aea85832c10f38999e3208802f8c78c64d287699b5565cca50523b7d294
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ef1f8eccf6f0195ee5c0b8a57c36f6448905d32bfc55a513d43fad2d1d747f5
-size 23178098
+oid sha256:23261c019d6dbf4a9bc3a482b8fcba0aab125d7510070fb37821c582073a7e58
+size 23106073

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 5000,
-    "exact_match": 62.5,
-    "f1": 72.40162540347497
+    "exact_match": 58.78,
+    "f1": 68.98441947756866
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.32572943477690974,
-    "train_runtime": 684.9677,
+    "epoch": 50.0,
+    "train_loss": 0.14661559768869908,
+    "train_runtime": 1698.5921,
     "train_samples": 5000,
-    "train_samples_per_second": 145.992,
-    "train_steps_per_second": 2.307
+    "train_samples_per_second": 147.181,
+    "train_steps_per_second": 2.325
 }

trainer_state.json

@@ -1,46 +1,70 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 20.0,
+  "epoch": 50.0,
   "eval_steps": 500,
-  "global_step": 1580,
+  "global_step": 3950,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 6.33,
-      "learning_rate": 3.4177215189873416e-05,
-      "loss": 0.9086,
+      "learning_rate": 4.367088607594937e-05,
+      "loss": 0.904,
       "step": 500
     },
     {
       "epoch": 12.66,
-      "learning_rate": 1.8354430379746836e-05,
-      "loss": 0.0941,
+      "learning_rate": 3.7341772151898736e-05,
+      "loss": 0.1028,
       "step": 1000
     },
     {
       "epoch": 18.99,
-      "learning_rate": 2.531645569620253e-06,
-      "loss": 0.0241,
+      "learning_rate": 3.10126582278481e-05,
+      "loss": 0.0435,
       "step": 1500
     },
     {
-      "epoch": 20.0,
-      "step": 1580,
-      "total_flos": 1.95972567552e+16,
-      "train_loss": 0.32572943477690974,
-      "train_runtime": 684.9677,
-      "train_samples_per_second": 145.992,
-      "train_steps_per_second": 2.307
+      "epoch": 25.32,
+      "learning_rate": 2.468354430379747e-05,
+      "loss": 0.0208,
+      "step": 2000
+    },
+    {
+      "epoch": 31.65,
+      "learning_rate": 1.8354430379746836e-05,
+      "loss": 0.0135,
+      "step": 2500
+    },
+    {
+      "epoch": 37.97,
+      "learning_rate": 1.2025316455696203e-05,
+      "loss": 0.0105,
+      "step": 3000
+    },
+    {
+      "epoch": 44.3,
+      "learning_rate": 5.69620253164557e-06,
+      "loss": 0.0188,
+      "step": 3500
+    },
+    {
+      "epoch": 50.0,
+      "step": 3950,
+      "total_flos": 4.8993141888e+16,
+      "train_loss": 0.14661559768869908,
+      "train_runtime": 1698.5921,
+      "train_samples_per_second": 147.181,
+      "train_steps_per_second": 2.325
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1580,
-  "num_train_epochs": 20,
+  "max_steps": 3950,
+  "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 1.95972567552e+16,
+  "total_flos": 4.8993141888e+16,
   "trial_name": null,
   "trial_params": null
 }