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": 51.56666666666667,
-    "f1": 61.20209643312287,
-    "train_loss": 0.5075189057985942,
-    "train_runtime": 413.5077,
+    "exact_match": 53.3,
+    "f1": 63.04894189678107,
+    "train_loss": 0.24828063900272052,
+    "train_runtime": 819.9397,
     "train_samples": 3000,
-    "train_samples_per_second": 362.75,
-    "train_steps_per_second": 1.451
+    "train_samples_per_second": 365.881,
+    "train_steps_per_second": 1.464
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f73842d78186e4627e15b8da431dc54bf9e092314127c8477eec236b8697cb6
+oid sha256:fec1e4a493250576b496ce9eb1e8ef1664b77c477a0aef901c3bd0e1e34d13b9
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2761a80eb4ea611c586d8610ae8ce9bd336d0306736189cb8cd0bb4c64dc8961
-size 14227253
+oid sha256:3bfa2f94dbfa1f1809eda0f1549268165ae69d4d98f3afddf526747a4e1e7ebf
+size 14255468

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 51.56666666666667,
-    "f1": 61.20209643312287
+    "exact_match": 53.3,
+    "f1": 63.04894189678107
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.5075189057985942,
-    "train_runtime": 413.5077,
+    "epoch": 100.0,
+    "train_loss": 0.24828063900272052,
+    "train_runtime": 819.9397,
     "train_samples": 3000,
-    "train_samples_per_second": 362.75,
-    "train_steps_per_second": 1.451
+    "train_samples_per_second": 365.881,
+    "train_steps_per_second": 1.464
 }

trainer_state.json

@@ -1,34 +1,40 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 600,
+  "global_step": 1200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 41.67,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.6047,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.5848,
       "step": 500
     },
     {
-      "epoch": 50.0,
-      "step": 600,
-      "total_flos": 1.46984737536e+16,
-      "train_loss": 0.5075189057985942,
-      "train_runtime": 413.5077,
-      "train_samples_per_second": 362.75,
-      "train_steps_per_second": 1.451
+      "epoch": 83.33,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0094,
+      "step": 1000
+    },
+    {
+      "epoch": 100.0,
+      "step": 1200,
+      "total_flos": 2.93969475072e+16,
+      "train_loss": 0.24828063900272052,
+      "train_runtime": 819.9397,
+      "train_samples_per_second": 365.881,
+      "train_steps_per_second": 1.464
     }
   ],
   "logging_steps": 500,
-  "max_steps": 600,
-  "num_train_epochs": 50,
+  "max_steps": 1200,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 1.46984737536e+16,
+  "total_flos": 2.93969475072e+16,
   "trial_name": null,
   "trial_params": null
 }