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": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 69.2,
-    "f1": 78.80268919426098,
-    "train_loss": 0.12083518410722414,
-    "train_runtime": 860.8664,
+    "exact_match": 64.5,
+    "f1": 75.09714036291676,
+    "train_loss": 0.06646478484074275,
+    "train_runtime": 1714.3663,
     "train_samples": 3000,
-    "train_samples_per_second": 174.243,
-    "train_steps_per_second": 1.394
+    "train_samples_per_second": 174.992,
+    "train_steps_per_second": 1.4
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e43ef52639ff48e0639cfd6126823b73cdc9112d105a29b9cb4ba112478e0d1f
+oid sha256:137642bddb3a9af2839a1635f3f57c46cb85cc522346443497e58fcbaf1f496f
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:efa0760e9ef81f16ff9ff760ffb1bcb891d7227278623cc03a7d7e3de6e8296c
-size 13512380
+oid sha256:8a0779ab49a33d1232abf63226048840adfdd0b6a0ab7510bc851ed64ff309b6
+size 13435214

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 50.0,
+    "epoch": 100.0,
     "eval_samples": 3000,
-    "exact_match": 69.2,
-    "f1": 78.80268919426098
+    "exact_match": 64.5,
+    "f1": 75.09714036291676
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 50.0,
-    "train_loss": 0.12083518410722414,
-    "train_runtime": 860.8664,
+    "epoch": 100.0,
+    "train_loss": 0.06646478484074275,
+    "train_runtime": 1714.3663,
     "train_samples": 3000,
-    "train_samples_per_second": 174.243,
-    "train_steps_per_second": 1.394
+    "train_samples_per_second": 174.992,
+    "train_steps_per_second": 1.4
 }

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": 1200,
+  "global_step": 2400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 20.83,
-      "learning_rate": 2.916666666666667e-05,
-      "loss": 0.2827,
+      "learning_rate": 3.958333333333333e-05,
+      "loss": 0.2896,
       "step": 500
     },
     {
       "epoch": 41.67,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.0068,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.0171,
       "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
+      "epoch": 62.5,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.0055,
+      "step": 1500
+    },
+    {
+      "epoch": 83.33,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0028,
+      "step": 2000
+    },
+    {
+      "epoch": 100.0,
+      "step": 2400,
+      "total_flos": 4968855705600000.0,
+      "train_loss": 0.06646478484074275,
+      "train_runtime": 1714.3663,
+      "train_samples_per_second": 174.992,
+      "train_steps_per_second": 1.4
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1200,
-  "num_train_epochs": 50,
+  "max_steps": 2400,
+  "num_train_epochs": 100,
   "save_steps": 500,
-  "total_flos": 2484427852800000.0,
+  "total_flos": 4968855705600000.0,
   "trial_name": null,
   "trial_params": null
 }