End of training

README.md

@@ -2,7 +2,10 @@
 license: apache-2.0
 base_model: bert-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 [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
+This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-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": 60.9,
-    "f1": 70.75629058027386,
-    "train_loss": 0.1784811174108627,
-    "train_runtime": 1041.8722,
+    "exact_match": 55.1,
+    "f1": 64.81720746704029,
+    "train_loss": 0.09985893122693326,
+    "train_runtime": 2075.477,
     "train_samples": 3000,
-    "train_samples_per_second": 143.972,
-    "train_steps_per_second": 2.256
+    "train_samples_per_second": 144.545,
+    "train_steps_per_second": 2.265
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a6e7b43a5ba3457d18b3017358485cc424e4f96e39c5d7db5cc082f3d1202cb
+oid sha256:1345a5fa88a2d2a8015bc84ce23c714da7b205bed4407eaef3112406987d6e36
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90791a77f163434b53738fe892c4c1336729e96370549c7167d9bdcecfc04615
-size 14171651
+oid sha256:ab5e68d818cf8e4632e6ed5e22386775642d2c032c9f0f4b10983e1340acfafa
+size 14061873

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 60.9,
-    "f1": 70.75629058027386
+    "exact_match": 55.1,
+    "f1": 64.81720746704029
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.1784811174108627,
-    "train_runtime": 1041.8722,
+    "epoch": 100.0,
+    "train_loss": 0.09985893122693326,
+    "train_runtime": 2075.477,
     "train_samples": 3000,
-    "train_samples_per_second": 143.972,
-    "train_steps_per_second": 2.256
+    "train_samples_per_second": 144.545,
+    "train_steps_per_second": 2.265
 }

trainer_state.json

@@ -1,52 +1,82 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 2350,
+  "global_step": 4700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 10.64,
-      "learning_rate": 3.936170212765958e-05,
-      "loss": 0.7712,
+      "learning_rate": 4.468085106382979e-05,
+      "loss": 0.77,
       "step": 500
     },
     {
       "epoch": 21.28,
-      "learning_rate": 2.8723404255319154e-05,
-      "loss": 0.0462,
+      "learning_rate": 3.936170212765958e-05,
+      "loss": 0.0564,
       "step": 1000
     },
     {
       "epoch": 31.91,
-      "learning_rate": 1.8085106382978724e-05,
-      "loss": 0.0141,
+      "learning_rate": 3.4042553191489365e-05,
+      "loss": 0.0198,
       "step": 1500
     },
     {
       "epoch": 42.55,
-      "learning_rate": 7.446808510638298e-06,
-      "loss": 0.005,
+      "learning_rate": 2.8723404255319154e-05,
+      "loss": 0.0113,
       "step": 2000
     },
     {
-      "epoch": 50.0,
-      "step": 2350,
-      "total_flos": 2.93958851328e+16,
-      "train_loss": 0.1784811174108627,
-      "train_runtime": 1041.8722,
-      "train_samples_per_second": 143.972,
-      "train_steps_per_second": 2.256
+      "epoch": 53.19,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.0061,
+      "step": 2500
+    },
+    {
+      "epoch": 63.83,
+      "learning_rate": 1.8085106382978724e-05,
+      "loss": 0.0063,
+      "step": 3000
+    },
+    {
+      "epoch": 74.47,
+      "learning_rate": 1.2765957446808511e-05,
+      "loss": 0.0077,
+      "step": 3500
+    },
+    {
+      "epoch": 85.11,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 0.0113,
+      "step": 4000
+    },
+    {
+      "epoch": 95.74,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 0.023,
+      "step": 4500
+    },
+    {
+      "epoch": 100.0,
+      "step": 4700,
+      "total_flos": 5.87917702656e+16,
+      "train_loss": 0.09985893122693326,
+      "train_runtime": 2075.477,
+      "train_samples_per_second": 144.545,
+      "train_steps_per_second": 2.265
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2350,
-  "num_train_epochs": 50,
+  "max_steps": 4700,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 2.93958851328e+16,
+  "total_flos": 5.87917702656e+16,
   "trial_name": null,
   "trial_params": null
 }