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": 50.0,
+    "epoch": 100.0,
     "eval_samples": 5000,
-    "exact_match": 59.5,
-    "f1": 68.69015298927383,
-    "train_loss": 0.1770759046228626,
-    "train_runtime": 1518.0723,
+    "exact_match": 59.14,
+    "f1": 68.39725861890774,
+    "train_loss": 0.09234336149843433,
+    "train_runtime": 3031.3949,
     "train_samples": 5000,
-    "train_samples_per_second": 164.683,
-    "train_steps_per_second": 2.602
+    "train_samples_per_second": 164.941,
+    "train_steps_per_second": 2.606
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48cfd77a45bfbe32525270b473a1cfde4850e3b6df6c6117ee735d0aa079cf79
+oid sha256:edab80ac1e569b85d29d002a99e6de8c755124ec91f0e901703789def845a89b
 size 5588

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2658a085001073c7efc9841c2a9ba931b1315c7e8d25fe0b069ec1e66227fec0
-size 23100491
+oid sha256:acd841d9f9d7ef5ef7467e4573de7d176da7b18a7214649bc38b791e795600e1
+size 23003836

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 5000,
-    "exact_match": 59.5,
-    "f1": 68.69015298927383
+    "exact_match": 59.14,
+    "f1": 68.39725861890774
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.1770759046228626,
-    "train_runtime": 1518.0723,
+    "epoch": 100.0,
+    "train_loss": 0.09234336149843433,
+    "train_runtime": 3031.3949,
     "train_samples": 5000,
-    "train_samples_per_second": 164.683,
-    "train_steps_per_second": 2.602
+    "train_samples_per_second": 164.941,
+    "train_steps_per_second": 2.606
 }

trainer_state.json

@@ -1,70 +1,118 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 3950,
+  "global_step": 7900,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 6.33,
-      "learning_rate": 4.367088607594937e-05,
-      "loss": 1.1777,
+      "learning_rate": 4.683544303797468e-05,
+      "loss": 1.1738,
       "step": 500
     },
     {
       "epoch": 12.66,
-      "learning_rate": 3.7341772151898736e-05,
-      "loss": 0.1233,
+      "learning_rate": 4.367088607594937e-05,
+      "loss": 0.1263,
       "step": 1000
     },
     {
       "epoch": 18.99,
-      "learning_rate": 3.10126582278481e-05,
-      "loss": 0.0499,
+      "learning_rate": 4.050632911392405e-05,
+      "loss": 0.0548,
       "step": 1500
     },
     {
       "epoch": 25.32,
-      "learning_rate": 2.468354430379747e-05,
-      "loss": 0.0223,
+      "learning_rate": 3.7341772151898736e-05,
+      "loss": 0.0291,
       "step": 2000
     },
     {
       "epoch": 31.65,
-      "learning_rate": 1.8354430379746836e-05,
-      "loss": 0.0116,
+      "learning_rate": 3.4177215189873416e-05,
+      "loss": 0.0192,
       "step": 2500
     },
     {
       "epoch": 37.97,
-      "learning_rate": 1.2025316455696203e-05,
-      "loss": 0.0068,
+      "learning_rate": 3.10126582278481e-05,
+      "loss": 0.0152,
       "step": 3000
     },
     {
       "epoch": 44.3,
-      "learning_rate": 5.69620253164557e-06,
-      "loss": 0.0047,
+      "learning_rate": 2.7848101265822786e-05,
+      "loss": 0.0102,
       "step": 3500
     },
     {
-      "epoch": 50.0,
-      "step": 3950,
-      "total_flos": 2.4497456256e+16,
-      "train_loss": 0.1770759046228626,
-      "train_runtime": 1518.0723,
-      "train_samples_per_second": 164.683,
-      "train_steps_per_second": 2.602
+      "epoch": 50.63,
+      "learning_rate": 2.468354430379747e-05,
+      "loss": 0.006,
+      "step": 4000
+    },
+    {
+      "epoch": 56.96,
+      "learning_rate": 2.1518987341772153e-05,
+      "loss": 0.0053,
+      "step": 4500
+    },
+    {
+      "epoch": 63.29,
+      "learning_rate": 1.8354430379746836e-05,
+      "loss": 0.0042,
+      "step": 5000
+    },
+    {
+      "epoch": 69.62,
+      "learning_rate": 1.5189873417721521e-05,
+      "loss": 0.0034,
+      "step": 5500
+    },
+    {
+      "epoch": 75.95,
+      "learning_rate": 1.2025316455696203e-05,
+      "loss": 0.003,
+      "step": 6000
+    },
+    {
+      "epoch": 82.28,
+      "learning_rate": 8.860759493670886e-06,
+      "loss": 0.0026,
+      "step": 6500
+    },
+    {
+      "epoch": 88.61,
+      "learning_rate": 5.69620253164557e-06,
+      "loss": 0.0025,
+      "step": 7000
+    },
+    {
+      "epoch": 94.94,
+      "learning_rate": 2.531645569620253e-06,
+      "loss": 0.0019,
+      "step": 7500
+    },
+    {
+      "epoch": 100.0,
+      "step": 7900,
+      "total_flos": 4.8994912512e+16,
+      "train_loss": 0.09234336149843433,
+      "train_runtime": 3031.3949,
+      "train_samples_per_second": 164.941,
+      "train_steps_per_second": 2.606
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3950,
-  "num_train_epochs": 50,
+  "max_steps": 7900,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 2.4497456256e+16,
+  "total_flos": 4.8994912512e+16,
   "trial_name": null,
   "trial_params": null
 }