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": 467.3287,
+    "exact_match": 53.9,
+    "f1": 63.80939869889286,
+    "train_loss": 0.16584412331382434,
+    "train_runtime": 937.6275,
     "train_samples": 3000,
-    "train_samples_per_second": 320.973,
-    "train_steps_per_second": 2.568
+    "train_samples_per_second": 319.956,
+    "train_steps_per_second": 2.56
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1cd8b4e728ae7aa107ddf3241a0f9e3daa3659d674a70b6e06b9c9d9f9cda88
+oid sha256:1da5a59ffa927e084b24841bcead5cbf422c42c00a8fb23076e8b6745adf8a62
 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:e3ee6b4cf816228ee0dd5e4f001644f39d555364bc5814ce35425d79186afd52
+size 14192461

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": 53.9,
+    "f1": 63.80939869889286
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.32951845953861875,
-    "train_runtime": 467.3287,
+    "epoch": 100.0,
+    "train_loss": 0.16584412331382434,
+    "train_runtime": 937.6275,
     "train_samples": 3000,
-    "train_samples_per_second": 320.973,
-    "train_steps_per_second": 2.568
+    "train_samples_per_second": 319.956,
+    "train_steps_per_second": 2.56
 }

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": 467.3287,
-      "train_samples_per_second": 320.973,
-      "train_steps_per_second": 2.568
+      "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": 0.16584412331382434,
+      "train_runtime": 937.6275,
+      "train_samples_per_second": 319.956,
+      "train_steps_per_second": 2.56
     }
   ],
   "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
 }