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": 68.86666666666666,
-    "f1": 77.76425661888325,
-    "train_loss": 0.09819502154986064,
-    "train_runtime": 4749.1078,
+    "exact_match": 63.4,
+    "f1": 73.77262733659492,
+    "train_loss": 1.6219618777011304,
+    "train_runtime": 165.6251,
     "train_samples": 3000,
-    "train_samples_per_second": 63.17,
-    "train_steps_per_second": 0.505
+    "train_samples_per_second": 54.34,
+    "train_steps_per_second": 0.851
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77d11f566704b8ea970a8b9270c4a9beee63dee776334b5e5fc069c5088dbf0c
+oid sha256:33ef665689044f6a1e66fbd5e20e9d9e121c1ce1f355282f6b6da4be33b5006c
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a5bfba2d642d62d6d3840b324816dc1169c1ca3f0e889b0b8dc530396b98917
-size 13896163
+oid sha256:a337395dabebd96d8258cb663727389215a2284f6326a9675fc999f71ddeda1a
+size 13841566

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 100.0,
+    "epoch": 3.0,
     "eval_samples": 3000,
-    "exact_match": 68.86666666666666,
-    "f1": 77.76425661888325
+    "exact_match": 63.4,
+    "f1": 73.77262733659492
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 100.0,
-    "train_loss": 0.09819502154986064,
-    "train_runtime": 4749.1078,
+    "epoch": 3.0,
+    "train_loss": 1.6219618777011304,
+    "train_runtime": 165.6251,
     "train_samples": 3000,
-    "train_samples_per_second": 63.17,
-    "train_steps_per_second": 0.505
+    "train_samples_per_second": 54.34,
+    "train_steps_per_second": 0.851
 }

trainer_state.json

@@ -1,52 +1,28 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 100.0,
+  "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 2400,
+  "global_step": 141,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 20.83,
-      "learning_rate": 3.958333333333333e-05,
-      "loss": 0.4465,
-      "step": 500
-    },
-    {
-      "epoch": 41.67,
-      "learning_rate": 2.916666666666667e-05,
-      "loss": 0.0159,
-      "step": 1000
-    },
-    {
-      "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
+      "epoch": 3.0,
+      "step": 141,
+      "total_flos": 6268772196864000.0,
+      "train_loss": 1.6219618777011304,
+      "train_runtime": 165.6251,
+      "train_samples_per_second": 54.34,
+      "train_steps_per_second": 0.851
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2400,
-  "num_train_epochs": 100,
+  "max_steps": 141,
+  "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 2.089590732288e+17,
+  "total_flos": 6268772196864000.0,
   "trial_name": null,
   "trial_params": null
 }