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": 3000,
-    "exact_match": 54.766666666666666,
-    "f1": 64.53016416256601,
-    "train_loss": 0.24304250493962715,
-    "train_runtime": 576.78,
+    "exact_match": 46.46666666666667,
+    "f1": 57.048156525818875,
+    "train_loss": 0.1373301104281811,
+    "train_runtime": 1170.2721,
     "train_samples": 3000,
-    "train_samples_per_second": 260.065,
-    "train_steps_per_second": 4.074
+    "train_samples_per_second": 256.351,
+    "train_steps_per_second": 4.016
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a222382bc3c5c55b27c6968574e5efec43b0d436b9660d08670cf2515dfa2323
+oid sha256:674b658cb5de3b2d73ab687b0015e16b01f592ed2c2a5e165578fb83c5f8c6fa
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:198cdcbb97d9181d7a77982d35961a64e1020a04c63338759614a03cf31ca153
-size 14142746
+oid sha256:31492035a229852f166abd6073c14daf95cfb304676e60079cee35e0ac6d87b9
+size 14143928

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 54.766666666666666,
-    "f1": 64.53016416256601
+    "exact_match": 46.46666666666667,
+    "f1": 57.048156525818875
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.24304250493962715,
-    "train_runtime": 576.78,
+    "epoch": 100.0,
+    "train_loss": 0.1373301104281811,
+    "train_runtime": 1170.2721,
     "train_samples": 3000,
-    "train_samples_per_second": 260.065,
-    "train_steps_per_second": 4.074
+    "train_samples_per_second": 256.351,
+    "train_steps_per_second": 4.016
 }

trainer_state.json

@@ -1,52 +1,82 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 2350,
+  "global_step": 4700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 10.64,
-      "learning_rate": 3.936170212765958e-05,
-      "loss": 1.0484,
+      "learning_rate": 4.468085106382979e-05,
+      "loss": 1.0426,
       "step": 500
     },
     {
       "epoch": 21.28,
-      "learning_rate": 2.8723404255319154e-05,
-      "loss": 0.0622,
+      "learning_rate": 3.936170212765958e-05,
+      "loss": 0.0678,
       "step": 1000
     },
     {
       "epoch": 31.91,
-      "learning_rate": 1.8085106382978724e-05,
-      "loss": 0.0191,
+      "learning_rate": 3.4042553191489365e-05,
+      "loss": 0.0241,
       "step": 1500
     },
     {
       "epoch": 42.55,
-      "learning_rate": 7.446808510638298e-06,
-      "loss": 0.0081,
+      "learning_rate": 2.8723404255319154e-05,
+      "loss": 0.0124,
       "step": 2000
     },
     {
-      "epoch": 50.0,
-      "step": 2350,
-      "total_flos": 1.46984737536e+16,
-      "train_loss": 0.24304250493962715,
-      "train_runtime": 576.78,
-      "train_samples_per_second": 260.065,
-      "train_steps_per_second": 4.074
+      "epoch": 53.19,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.0083,
+      "step": 2500
+    },
+    {
+      "epoch": 63.83,
+      "learning_rate": 1.8085106382978724e-05,
+      "loss": 0.0099,
+      "step": 3000
+    },
+    {
+      "epoch": 74.47,
+      "learning_rate": 1.2765957446808511e-05,
+      "loss": 0.0117,
+      "step": 3500
+    },
+    {
+      "epoch": 85.11,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 0.0168,
+      "step": 4000
+    },
+    {
+      "epoch": 95.74,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 0.0463,
+      "step": 4500
+    },
+    {
+      "epoch": 100.0,
+      "step": 4700,
+      "total_flos": 2.93969475072e+16,
+      "train_loss": 0.1373301104281811,
+      "train_runtime": 1170.2721,
+      "train_samples_per_second": 256.351,
+      "train_steps_per_second": 4.016
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2350,
-  "num_train_epochs": 50,
+  "max_steps": 4700,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 1.46984737536e+16,
+  "total_flos": 2.93969475072e+16,
   "trial_name": null,
   "trial_params": null
 }