End of training

README.md

@@ -2,7 +2,10 @@
 license: apache-2.0
 base_model: albert-base-v2
 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 [albert-base-v2](https://huggingface.co/albert-base-v2) on the None dataset.
+This model is a fine-tuned version of [albert-base-v2](https://huggingface.co/albert-base-v2) on the squad dataset.
 
 ## Model description
 

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 68.6,
-    "f1": 78.54201655265396,
-    "train_loss": 0.27897618611653646,
-    "train_runtime": 346.2205,
+    "exact_match": 69.2,
+    "f1": 78.80268919426098,
+    "train_loss": 0.12083518410722414,
+    "train_runtime": 860.8664,
     "train_samples": 3000,
-    "train_samples_per_second": 173.3,
-    "train_steps_per_second": 1.386
+    "train_samples_per_second": 174.243,
+    "train_steps_per_second": 1.394
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d916229e58bdb74b3c2977044d40c40f212fb01afcc0daa41178ed3766767729
+oid sha256:e43ef52639ff48e0639cfd6126823b73cdc9112d105a29b9cb4ba112478e0d1f
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d20435e70728505469cfc37471c86151003985b26493c2664a2e544dc3850b9
-size 13559975
+oid sha256:efa0760e9ef81f16ff9ff760ffb1bcb891d7227278623cc03a7d7e3de6e8296c
+size 13512380

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 68.6,
-    "f1": 78.54201655265396
+    "exact_match": 69.2,
+    "f1": 78.80268919426098
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.27897618611653646,
-    "train_runtime": 346.2205,
+    "epoch": 50.0,
+    "train_loss": 0.12083518410722414,
+    "train_runtime": 860.8664,
     "train_samples": 3000,
-    "train_samples_per_second": 173.3,
-    "train_steps_per_second": 1.386
+    "train_samples_per_second": 174.243,
+    "train_steps_per_second": 1.394
 }

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": 993771141120000.0,
-      "train_loss": 0.27897618611653646,
-      "train_runtime": 346.2205,
-      "train_samples_per_second": 173.3,
-      "train_steps_per_second": 1.386
+      "epoch": 20.83,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.2827,
+      "step": 500
+    },
+    {
+      "epoch": 41.67,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0068,
+      "step": 1000
+    },
+    {
+      "epoch": 50.0,
+      "step": 1200,
+      "total_flos": 2484427852800000.0,
+      "train_loss": 0.12083518410722414,
+      "train_runtime": 860.8664,
+      "train_samples_per_second": 174.243,
+      "train_steps_per_second": 1.394
     }
   ],
   "logging_steps": 500,
-  "max_steps": 480,
-  "num_train_epochs": 20,
+  "max_steps": 1200,
+  "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 993771141120000.0,
+  "total_flos": 2484427852800000.0,
   "trial_name": null,
   "trial_params": null
 }