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": 50.96666666666667,
-    "f1": 61.039962816049005,
-    "train_loss": 0.8416755040486653,
-    "train_runtime": 189.9259,
+    "exact_match": 54.56666666666667,
+    "f1": 64.45574575428589,
+    "train_loss": 0.32951845953861875,
+    "train_runtime": 467.3287,
     "train_samples": 3000,
-    "train_samples_per_second": 315.913,
-    "train_steps_per_second": 2.527
+    "train_samples_per_second": 320.973,
+    "train_steps_per_second": 2.568
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f016482b682c64c48d5d65758b22a28315d51b380c55d06f2c0340a12b1177ae
+oid sha256:d1cd8b4e728ae7aa107ddf3241a0f9e3daa3659d674a70b6e06b9c9d9f9cda88
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f18572031ea69d5c57308129454c4bff9039f881c9f29ed7362e13ca924c6e6a
-size 14227015
+oid sha256:d16c3e2bdcc66f14e12e92a13b5ad893a78fe7e79c7edff7a11b0dda83b1de4e
+size 14249017

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 50.96666666666667,
-    "f1": 61.039962816049005
+    "exact_match": 54.56666666666667,
+    "f1": 64.45574575428589
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.8416755040486653,
-    "train_runtime": 189.9259,
+    "epoch": 50.0,
+    "train_loss": 0.32951845953861875,
+    "train_runtime": 467.3287,
     "train_samples": 3000,
-    "train_samples_per_second": 315.913,
-    "train_steps_per_second": 2.527
+    "train_samples_per_second": 320.973,
+    "train_steps_per_second": 2.568
 }

trainer_state.json

@@ -1,28 +1,40 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 20.0,
+  "epoch": 50.0,
   "eval_steps": 500,
-  "global_step": 480,
+  "global_step": 1200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 20.0,
-      "step": 480,
-      "total_flos": 5879389501440000.0,
-      "train_loss": 0.8416755040486653,
-      "train_runtime": 189.9259,
-      "train_samples_per_second": 315.913,
-      "train_steps_per_second": 2.527
+      "epoch": 20.83,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.7659,
+      "step": 500
+    },
+    {
+      "epoch": 41.67,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0219,
+      "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
     }
   ],
   "logging_steps": 500,
-  "max_steps": 480,
-  "num_train_epochs": 20,
+  "max_steps": 1200,
+  "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 5879389501440000.0,
+  "total_flos": 1.46984737536e+16,
   "trial_name": null,
   "trial_params": null
 }