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": 57.58,
-    "f1": 67.21309314821316,
-    "train_loss": 0.24194956755638122,
-    "train_runtime": 1318.7789,
+    "exact_match": 58.36,
+    "f1": 67.63478561868162,
+    "train_loss": 0.12255857245624066,
+    "train_runtime": 2635.091,
     "train_samples": 5000,
-    "train_samples_per_second": 189.569,
-    "train_steps_per_second": 1.517
+    "train_samples_per_second": 189.747,
+    "train_steps_per_second": 1.518
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:34f9c8c6b81e6c51ed1bf689bcd7d9afa7d549c13aab8759b5293a81f0562e75
+oid sha256:5bd04ab629d9115ddb3c5fb00216165f5ed1049866e5870faf646b7b757f5a85
 size 5588

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7ba5ce8ae15e76967b66e6b24195f351bd78ffc0bc318203bd3d7cfa9805ca0
-size 23113024
+oid sha256:bcc2b3dc394ff1ea957b66d8672e17eca66dd6f87a6fe6c047ce62debbe5cdeb
+size 23010977

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 5000,
-    "exact_match": 57.58,
-    "f1": 67.21309314821316
+    "exact_match": 58.36,
+    "f1": 67.63478561868162
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.24194956755638122,
-    "train_runtime": 1318.7789,
+    "epoch": 100.0,
+    "train_loss": 0.12255857245624066,
+    "train_runtime": 2635.091,
     "train_samples": 5000,
-    "train_samples_per_second": 189.569,
-    "train_steps_per_second": 1.517
+    "train_samples_per_second": 189.747,
+    "train_steps_per_second": 1.518
 }

trainer_state.json

@@ -1,52 +1,76 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 2000,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 12.5,
-      "learning_rate": 3.7500000000000003e-05,
-      "loss": 0.8888,
+      "learning_rate": 4.375e-05,
+      "loss": 0.8815,
       "step": 500
     },
     {
       "epoch": 25.0,
-      "learning_rate": 2.5e-05,
+      "learning_rate": 3.7500000000000003e-05,
       "loss": 0.0559,
       "step": 1000
     },
     {
       "epoch": 37.5,
-      "learning_rate": 1.25e-05,
-      "loss": 0.0163,
+      "learning_rate": 3.125e-05,
+      "loss": 0.0193,
       "step": 1500
     },
     {
       "epoch": 50.0,
-      "learning_rate": 0.0,
-      "loss": 0.0068,
+      "learning_rate": 2.5e-05,
+      "loss": 0.009,
       "step": 2000
     },
     {
-      "epoch": 50.0,
-      "step": 2000,
-      "total_flos": 2.4497456256e+16,
-      "train_loss": 0.24194956755638122,
-      "train_runtime": 1318.7789,
-      "train_samples_per_second": 189.569,
-      "train_steps_per_second": 1.517
+      "epoch": 62.5,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.0058,
+      "step": 2500
+    },
+    {
+      "epoch": 75.0,
+      "learning_rate": 1.25e-05,
+      "loss": 0.0042,
+      "step": 3000
+    },
+    {
+      "epoch": 87.5,
+      "learning_rate": 6.25e-06,
+      "loss": 0.0027,
+      "step": 3500
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 0.0,
+      "loss": 0.002,
+      "step": 4000
+    },
+    {
+      "epoch": 100.0,
+      "step": 4000,
+      "total_flos": 4.8994912512e+16,
+      "train_loss": 0.12255857245624066,
+      "train_runtime": 2635.091,
+      "train_samples_per_second": 189.747,
+      "train_steps_per_second": 1.518
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2000,
-  "num_train_epochs": 50,
+  "max_steps": 4000,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 2.4497456256e+16,
+  "total_flos": 4.8994912512e+16,
   "trial_name": null,
   "trial_params": null
 }