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": 54.56666666666667,
-    "f1": 64.45574575428589,
-    "train_loss": 0.32951845953861875,
-    "train_runtime": 470.6172,
+    "exact_match": 54.1,
+    "f1": 63.82638623890578,
+    "train_loss": 0.16588029886285463,
+    "train_runtime": 939.8165,
     "train_samples": 3000,
-    "train_samples_per_second": 318.73,
-    "train_steps_per_second": 2.55
+    "train_samples_per_second": 319.211,
+    "train_steps_per_second": 2.554
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7231e37486541e75ad1748ebfdb3b7acea5d6a7fb1c3b4feeffed9d63c75b668
+oid sha256:f09f14ebf9eca2ef64d0a01145d9b973a30cfeaca06f0582baa8951d9822eae1
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d16c3e2bdcc66f14e12e92a13b5ad893a78fe7e79c7edff7a11b0dda83b1de4e
-size 14249017
+oid sha256:0f4b631b438c1d93043e1093c2c9084f751876e8b810c62d303137b593f3284a
+size 14170149

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 54.56666666666667,
-    "f1": 64.45574575428589
+    "exact_match": 54.1,
+    "f1": 63.82638623890578
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.32951845953861875,
-    "train_runtime": 470.6172,
+    "epoch": 100.0,
+    "train_loss": 0.16588029886285463,
+    "train_runtime": 939.8165,
     "train_samples": 3000,
-    "train_samples_per_second": 318.73,
-    "train_steps_per_second": 2.55
+    "train_samples_per_second": 319.211,
+    "train_steps_per_second": 2.554
 }

trainer_state.json

@@ -1,40 +1,52 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 1200,
+  "global_step": 2400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 20.83,
-      "learning_rate": 2.916666666666667e-05,
-      "loss": 0.7659,
+      "learning_rate": 3.958333333333333e-05,
+      "loss": 0.7576,
       "step": 500
     },
     {
       "epoch": 41.67,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.0219,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.0249,
       "step": 1000
     },
     {
-      "epoch": 50.0,
-      "step": 1200,
-      "total_flos": 1.46984737536e+16,
-      "train_loss": 0.32951845953861875,
-      "train_runtime": 470.6172,
-      "train_samples_per_second": 318.73,
-      "train_steps_per_second": 2.55
+      "epoch": 62.5,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.0074,
+      "step": 1500
+    },
+    {
+      "epoch": 83.33,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0038,
+      "step": 2000
+    },
+    {
+      "epoch": 100.0,
+      "step": 2400,
+      "total_flos": 2.93969475072e+16,
+      "train_loss": 0.16588029886285463,
+      "train_runtime": 939.8165,
+      "train_samples_per_second": 319.211,
+      "train_steps_per_second": 2.554
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1200,
-  "num_train_epochs": 50,
+  "max_steps": 2400,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 1.46984737536e+16,
+  "total_flos": 2.93969475072e+16,
   "trial_name": null,
   "trial_params": null
 }