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": 56.36666666666667,
-    "f1": 65.6870988663549,
-    "train_loss": 0.4278867336029702,
-    "train_runtime": 303.7628,
+    "exact_match": 47.233333333333334,
+    "f1": 57.723557986903565,
+    "train_loss": 0.1978351655919501,
+    "train_runtime": 783.094,
     "train_samples": 3000,
-    "train_samples_per_second": 197.523,
-    "train_steps_per_second": 6.189
+    "train_samples_per_second": 191.548,
+    "train_steps_per_second": 6.002
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:595225b3aa7c124c5f885a052fc420cdeab0cde9c0f9c0f591a4d5c26d64c460
+oid sha256:fbd2d598f9b28d49a56ad7856529421c5ad5920bc930c02adeb9ff43faaf2126
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e8750c3930348fa2231008fbd0cd44e4a8f1b07f3ca07a864dbe897357f9d11
-size 14240855
+oid sha256:dfd33d0496ad36f95724c9f4b0cad861fbdfd34b4d7adc0c5afb9254e6f92b45
+size 14129439

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 56.36666666666667,
-    "f1": 65.6870988663549
+    "exact_match": 47.233333333333334,
+    "f1": 57.723557986903565
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.4278867336029702,
-    "train_runtime": 303.7628,
+    "epoch": 50.0,
+    "train_loss": 0.1978351655919501,
+    "train_runtime": 783.094,
     "train_samples": 3000,
-    "train_samples_per_second": 197.523,
-    "train_steps_per_second": 6.189
+    "train_samples_per_second": 191.548,
+    "train_steps_per_second": 6.002
 }

trainer_state.json

@@ -1,46 +1,82 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 20.0,
+  "epoch": 50.0,
   "eval_steps": 500,
-  "global_step": 1880,
+  "global_step": 4700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 5.32,
-      "learning_rate": 3.670212765957447e-05,
-      "loss": 1.4055,
+      "learning_rate": 4.468085106382979e-05,
+      "loss": 1.3964,
       "step": 500
     },
     {
       "epoch": 10.64,
-      "learning_rate": 2.340425531914894e-05,
-      "loss": 0.1477,
+      "learning_rate": 3.936170212765958e-05,
+      "loss": 0.1539,
       "step": 1000
     },
     {
       "epoch": 15.96,
-      "learning_rate": 1.0106382978723404e-05,
-      "loss": 0.0432,
+      "learning_rate": 3.4042553191489365e-05,
+      "loss": 0.0614,
       "step": 1500
     },
     {
-      "epoch": 20.0,
-      "step": 1880,
-      "total_flos": 5879389501440000.0,
-      "train_loss": 0.4278867336029702,
-      "train_runtime": 303.7628,
-      "train_samples_per_second": 197.523,
-      "train_steps_per_second": 6.189
+      "epoch": 21.28,
+      "learning_rate": 2.8723404255319154e-05,
+      "loss": 0.0292,
+      "step": 2000
+    },
+    {
+      "epoch": 26.6,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.0187,
+      "step": 2500
+    },
+    {
+      "epoch": 31.91,
+      "learning_rate": 1.8085106382978724e-05,
+      "loss": 0.0171,
+      "step": 3000
+    },
+    {
+      "epoch": 37.23,
+      "learning_rate": 1.2765957446808511e-05,
+      "loss": 0.0174,
+      "step": 3500
+    },
+    {
+      "epoch": 42.55,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 0.0293,
+      "step": 4000
+    },
+    {
+      "epoch": 47.87,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 0.0693,
+      "step": 4500
+    },
+    {
+      "epoch": 50.0,
+      "step": 4700,
+      "total_flos": 1.46984737536e+16,
+      "train_loss": 0.1978351655919501,
+      "train_runtime": 783.094,
+      "train_samples_per_second": 191.548,
+      "train_steps_per_second": 6.002
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1880,
-  "num_train_epochs": 20,
+  "max_steps": 4700,
+  "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 5879389501440000.0,
+  "total_flos": 1.46984737536e+16,
   "trial_name": null,
   "trial_params": null
 }