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": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 56.46666666666667,
-    "f1": 65.69798974381443,
-    "train_loss": 0.4279018166217398,
-    "train_runtime": 303.9253,
+    "exact_match": 37.833333333333336,
+    "f1": 49.61991770612217,
+    "train_loss": 0.2530505502984879,
+    "train_runtime": 753.9352,
     "train_samples": 3000,
-    "train_samples_per_second": 197.417,
-    "train_steps_per_second": 6.186
+    "train_samples_per_second": 198.956,
+    "train_steps_per_second": 6.234
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b05c8907dfc280bd1c1e1e817267322e98030804d0488d1fdd3fb33e2b44b7fe
+oid sha256:bdf89c5a0521e037b3c80248a874cc24d5f7b9f10473d0d98a25dcd76e0ea2fc
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd412c736e41e466d98fd5b5e4084e176a1136bfb43b240de8217cee79037206
-size 14228580
+oid sha256:4cf0dd18818fa5f8dc24f97c37c1ea720b7dbe3c0a61fc76c2cc4b72527f3791
+size 14182852

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 56.46666666666667,
-    "f1": 65.69798974381443
+    "exact_match": 37.833333333333336,
+    "f1": 49.61991770612217
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.4279018166217398,
-    "train_runtime": 303.9253,
+    "epoch": 50.0,
+    "train_loss": 0.2530505502984879,
+    "train_runtime": 753.9352,
     "train_samples": 3000,
-    "train_samples_per_second": 197.417,
-    "train_steps_per_second": 6.186
+    "train_samples_per_second": 198.956,
+    "train_steps_per_second": 6.234
 }

trainer_state.json

@@ -1,46 +1,82 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 20.0,
+  "epoch": 50.0,
   "eval_steps": 500,
-  "global_step": 1880,
+  "global_step": 4700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 5.32,
-      "learning_rate": 3.670212765957447e-05,
-      "loss": 1.4055,
+      "learning_rate": 4.468085106382979e-05,
+      "loss": 1.3964,
       "step": 500
     },
     {
       "epoch": 10.64,
-      "learning_rate": 2.340425531914894e-05,
-      "loss": 0.1477,
+      "learning_rate": 3.936170212765958e-05,
+      "loss": 0.1539,
       "step": 1000
     },
     {
       "epoch": 15.96,
-      "learning_rate": 1.0106382978723404e-05,
-      "loss": 0.0432,
+      "learning_rate": 3.4042553191489365e-05,
+      "loss": 0.0614,
       "step": 1500
     },
     {
-      "epoch": 20.0,
-      "step": 1880,
-      "total_flos": 5879389501440000.0,
-      "train_loss": 0.4279018166217398,
-      "train_runtime": 303.9253,
-      "train_samples_per_second": 197.417,
-      "train_steps_per_second": 6.186
+      "epoch": 21.28,
+      "learning_rate": 2.8723404255319154e-05,
+      "loss": 0.0293,
+      "step": 2000
+    },
+    {
+      "epoch": 26.6,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.0184,
+      "step": 2500
+    },
+    {
+      "epoch": 31.91,
+      "learning_rate": 1.8085106382978724e-05,
+      "loss": 0.0199,
+      "step": 3000
+    },
+    {
+      "epoch": 37.23,
+      "learning_rate": 1.2765957446808511e-05,
+      "loss": 0.0284,
+      "step": 3500
+    },
+    {
+      "epoch": 42.55,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 0.0719,
+      "step": 4000
+    },
+    {
+      "epoch": 47.87,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 0.2953,
+      "step": 4500
+    },
+    {
+      "epoch": 50.0,
+      "step": 4700,
+      "total_flos": 1.46984737536e+16,
+      "train_loss": 0.2530505502984879,
+      "train_runtime": 753.9352,
+      "train_samples_per_second": 198.956,
+      "train_steps_per_second": 6.234
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1880,
-  "num_train_epochs": 20,
+  "max_steps": 4700,
+  "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 5879389501440000.0,
+  "total_flos": 1.46984737536e+16,
   "trial_name": null,
   "trial_params": null
 }