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": 5000,
-    "exact_match": 55.46,
-    "f1": 65.39732707076156,
-    "train_loss": 0.353411434173584,
-    "train_runtime": 1204.7422,
+    "exact_match": 56.88,
+    "f1": 66.48020899451993,
+    "train_loss": 0.17576227641105652,
+    "train_runtime": 2398.3247,
     "train_samples": 5000,
-    "train_samples_per_second": 207.513,
-    "train_steps_per_second": 0.83
+    "train_samples_per_second": 208.479,
+    "train_steps_per_second": 0.834
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d22e13f9bd25a936e18d19330fbee7c8e0cac9fac2d3233ad28be8fd26bda883
+oid sha256:ec4a23c63b15f4ae27d7b33578d87ab01862c7fa53141230d6aae054fa06548f
 size 5588

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05d174312fc5b39c480130e517103327ac794e5a1a97b52c0668bd85017ef653
-size 23178091
+oid sha256:78f03378522ffff9b7fb48dc04e4150a5e9eee4f22dc91364c7ac785e9c9791f
+size 23161247

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 5000,
-    "exact_match": 55.46,
-    "f1": 65.39732707076156
+    "exact_match": 56.88,
+    "f1": 66.48020899451993
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.353411434173584,
-    "train_runtime": 1204.7422,
+    "epoch": 100.0,
+    "train_loss": 0.17576227641105652,
+    "train_runtime": 2398.3247,
     "train_samples": 5000,
-    "train_samples_per_second": 207.513,
-    "train_steps_per_second": 0.83
+    "train_samples_per_second": 208.479,
+    "train_steps_per_second": 0.834
 }

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": 1000,
+  "global_step": 2000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 25.0,
-      "learning_rate": 2.5e-05,
-      "loss": 0.6839,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.6704,
       "step": 500
     },
     {
       "epoch": 50.0,
-      "learning_rate": 0.0,
-      "loss": 0.0229,
+      "learning_rate": 2.5e-05,
+      "loss": 0.0221,
       "step": 1000
     },
     {
-      "epoch": 50.0,
-      "step": 1000,
-      "total_flos": 2.4497456256e+16,
-      "train_loss": 0.353411434173584,
-      "train_runtime": 1204.7422,
-      "train_samples_per_second": 207.513,
-      "train_steps_per_second": 0.83
+      "epoch": 75.0,
+      "learning_rate": 1.25e-05,
+      "loss": 0.0071,
+      "step": 1500
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 0.0,
+      "loss": 0.0035,
+      "step": 2000
+    },
+    {
+      "epoch": 100.0,
+      "step": 2000,
+      "total_flos": 4.8994912512e+16,
+      "train_loss": 0.17576227641105652,
+      "train_runtime": 2398.3247,
+      "train_samples_per_second": 208.479,
+      "train_steps_per_second": 0.834
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1000,
-  "num_train_epochs": 50,
+  "max_steps": 2000,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 2.4497456256e+16,
+  "total_flos": 4.8994912512e+16,
   "trial_name": null,
   "trial_params": null
 }