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": 5000,
-    "exact_match": 63.06,
-    "f1": 72.7342051521597,
-    "train_loss": 0.18040521144866944,
-    "train_runtime": 1459.0895,
+    "exact_match": 59.0,
+    "f1": 68.71097105662648,
+    "train_loss": 0.09883097648620605,
+    "train_runtime": 2920.5642,
     "train_samples": 5000,
-    "train_samples_per_second": 171.34,
-    "train_steps_per_second": 1.371
+    "train_samples_per_second": 171.2,
+    "train_steps_per_second": 1.37
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60dad82c8d2c07d6638e17b1fbcbce92421798356ab3901e65095c21526f4a73
+oid sha256:f09320c21c448f4f2ed9cf7d39d087b2c4fe9d110fa9270c6298dcd4d4009258
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3723ef2dd87dd70451ba04c4577b173f0082f70b64d2ef9f9e5c95918df25a8d
-size 23207852
+oid sha256:7f96c111c69214a4ef52204d906955e559d65a293df6fdb935c51edfd7618667
+size 22981173

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 5000,
-    "exact_match": 63.06,
-    "f1": 72.7342051521597
+    "exact_match": 59.0,
+    "f1": 68.71097105662648
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.18040521144866944,
-    "train_runtime": 1459.0895,
+    "epoch": 100.0,
+    "train_loss": 0.09883097648620605,
+    "train_runtime": 2920.5642,
     "train_samples": 5000,
-    "train_samples_per_second": 171.34,
-    "train_steps_per_second": 1.371
+    "train_samples_per_second": 171.2,
+    "train_steps_per_second": 1.37
 }

trainer_state.json

@@ -1,52 +1,76 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 50.0,
+  "epoch": 100.0,
   "eval_steps": 500,
-  "global_step": 2000,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 12.5,
-      "learning_rate": 3.7500000000000003e-05,
-      "loss": 0.6645,
+      "learning_rate": 4.375e-05,
+      "loss": 0.6616,
       "step": 500
     },
     {
       "epoch": 25.0,
-      "learning_rate": 2.5e-05,
-      "loss": 0.0397,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.0458,
       "step": 1000
     },
     {
       "epoch": 37.5,
-      "learning_rate": 1.25e-05,
-      "loss": 0.0128,
+      "learning_rate": 3.125e-05,
+      "loss": 0.0179,
       "step": 1500
     },
     {
       "epoch": 50.0,
-      "learning_rate": 0.0,
-      "loss": 0.0046,
+      "learning_rate": 2.5e-05,
+      "loss": 0.009,
       "step": 2000
     },
     {
-      "epoch": 50.0,
-      "step": 2000,
-      "total_flos": 4.8993141888e+16,
-      "train_loss": 0.18040521144866944,
-      "train_runtime": 1459.0895,
-      "train_samples_per_second": 171.34,
-      "train_steps_per_second": 1.371
+      "epoch": 62.5,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.0064,
+      "step": 2500
+    },
+    {
+      "epoch": 75.0,
+      "learning_rate": 1.25e-05,
+      "loss": 0.0077,
+      "step": 3000
+    },
+    {
+      "epoch": 87.5,
+      "learning_rate": 6.25e-06,
+      "loss": 0.0104,
+      "step": 3500
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 0.0,
+      "loss": 0.0319,
+      "step": 4000
+    },
+    {
+      "epoch": 100.0,
+      "step": 4000,
+      "total_flos": 9.7986283776e+16,
+      "train_loss": 0.09883097648620605,
+      "train_runtime": 2920.5642,
+      "train_samples_per_second": 171.2,
+      "train_steps_per_second": 1.37
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2000,
-  "num_train_epochs": 50,
+  "max_steps": 4000,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 4.8993141888e+16,
+  "total_flos": 9.7986283776e+16,
   "trial_name": null,
   "trial_params": null
 }