End of training

README.md

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

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 5.0,
+    "epoch": 10.0,
     "eval_samples": 10784,
-    "exact_match": 76.12109744560075,
-    "f1": 84.67154895231297,
-    "train_loss": 1.0621621633540688,
-    "train_runtime": 1213.6111,
+    "exact_match": 72.26111636707664,
+    "f1": 81.62230488906296,
+    "train_loss": 0.7383389687951589,
+    "train_runtime": 2418.3642,
     "train_samples": 88524,
-    "train_samples_per_second": 364.713,
-    "train_steps_per_second": 1.425
+    "train_samples_per_second": 366.049,
+    "train_steps_per_second": 1.431
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:483692591b39cae794e2bf79a7e2817aa98b0b93c741e4ce67735869612b26ce
+oid sha256:5e6392f2477459d5693b48db94ee43d9a1d0d90747dc005d9505d5d6aa8358fb
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd4faa04e2b0b10a19c8b25a1984c82001237cf0d28f33b09c126db7276f634b
-size 49324536
+oid sha256:fcd504c9f2dc1971ef98521e18441e60758e1e762b3444e5ae615b51f44ceb5f
+size 49254973

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 5.0,
+    "epoch": 10.0,
     "eval_samples": 10784,
-    "exact_match": 76.12109744560075,
-    "f1": 84.67154895231297
+    "exact_match": 72.26111636707664,
+    "f1": 81.62230488906296
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 5.0,
-    "train_loss": 1.0621621633540688,
-    "train_runtime": 1213.6111,
+    "epoch": 10.0,
+    "train_loss": 0.7383389687951589,
+    "train_runtime": 2418.3642,
     "train_samples": 88524,
-    "train_samples_per_second": 364.713,
-    "train_steps_per_second": 1.425
+    "train_samples_per_second": 366.049,
+    "train_steps_per_second": 1.431
 }

trainer_state.json

@@ -1,46 +1,64 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 5.0,
+  "epoch": 10.0,
   "eval_steps": 500,
-  "global_step": 1730,
+  "global_step": 3460,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.45,
-      "learning_rate": 3.554913294797688e-05,
-      "loss": 1.6738,
+      "learning_rate": 4.2774566473988445e-05,
+      "loss": 1.6654,
       "step": 500
     },
     {
       "epoch": 2.89,
-      "learning_rate": 2.1098265895953757e-05,
-      "loss": 0.9544,
+      "learning_rate": 3.554913294797688e-05,
+      "loss": 0.9341,
       "step": 1000
     },
     {
       "epoch": 4.34,
-      "learning_rate": 6.647398843930635e-06,
-      "loss": 0.7396,
+      "learning_rate": 2.832369942196532e-05,
+      "loss": 0.6779,
       "step": 1500
     },
     {
-      "epoch": 5.0,
-      "step": 1730,
-      "total_flos": 4.337225635212288e+16,
-      "train_loss": 1.0621621633540688,
-      "train_runtime": 1213.6111,
-      "train_samples_per_second": 364.713,
-      "train_steps_per_second": 1.425
+      "epoch": 5.78,
+      "learning_rate": 2.1098265895953757e-05,
+      "loss": 0.5258,
+      "step": 2000
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 1.3872832369942197e-05,
+      "loss": 0.4303,
+      "step": 2500
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 6.647398843930635e-06,
+      "loss": 0.4153,
+      "step": 3000
+    },
+    {
+      "epoch": 10.0,
+      "step": 3460,
+      "total_flos": 8.674451270424576e+16,
+      "train_loss": 0.7383389687951589,
+      "train_runtime": 2418.3642,
+      "train_samples_per_second": 366.049,
+      "train_steps_per_second": 1.431
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1730,
-  "num_train_epochs": 5,
+  "max_steps": 3460,
+  "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 4.337225635212288e+16,
+  "total_flos": 8.674451270424576e+16,
   "trial_name": null,
   "trial_params": null
 }