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": 67.6,
-    "f1": 77.98001410061255,
-    "train_loss": 0.20614639479429164,
-    "train_runtime": 429.6446,
+    "exact_match": 36.03333333333333,
+    "f1": 48.263453219658174,
+    "train_loss": 0.2573057787469093,
+    "train_runtime": 1080.3341,
     "train_samples": 3000,
-    "train_samples_per_second": 139.65,
-    "train_steps_per_second": 4.376
+    "train_samples_per_second": 138.846,
+    "train_steps_per_second": 4.351
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b3207b354104ae2aa7322b0e77c14a354007e7a43bd391df0c632e053c8bfd7
+oid sha256:78ebfbb95cb48deeaf9e5d6ab394813ee2be091b59b5813efedd3396e672d80e
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed67c57fcb4a38c9f758f5a71a7447f48c31bb3dd85f06aff1a867da2121d29e
-size 13489570
+oid sha256:016a1097f3c389b69090372f4e9e5a10876bbfdcb7eb3bad7d1c5556d3a0af46
+size 13503411

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 67.6,
-    "f1": 77.98001410061255
+    "exact_match": 36.03333333333333,
+    "f1": 48.263453219658174
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.20614639479429164,
-    "train_runtime": 429.6446,
+    "epoch": 50.0,
+    "train_loss": 0.2573057787469093,
+    "train_runtime": 1080.3341,
     "train_samples": 3000,
-    "train_samples_per_second": 139.65,
-    "train_steps_per_second": 4.376
+    "train_samples_per_second": 138.846,
+    "train_steps_per_second": 4.351
 }

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": 0.6754,
+      "learning_rate": 4.468085106382979e-05,
+      "loss": 0.6855,
       "step": 500
     },
     {
       "epoch": 10.64,
-      "learning_rate": 2.340425531914894e-05,
-      "loss": 0.0842,
+      "learning_rate": 3.936170212765958e-05,
+      "loss": 0.1069,
       "step": 1000
     },
     {
       "epoch": 15.96,
-      "learning_rate": 1.0106382978723404e-05,
-      "loss": 0.0137,
+      "learning_rate": 3.4042553191489365e-05,
+      "loss": 0.0548,
       "step": 1500
     },
     {
-      "epoch": 20.0,
-      "step": 1880,
-      "total_flos": 993771141120000.0,
-      "train_loss": 0.20614639479429164,
-      "train_runtime": 429.6446,
-      "train_samples_per_second": 139.65,
-      "train_steps_per_second": 4.376
+      "epoch": 21.28,
+      "learning_rate": 2.8723404255319154e-05,
+      "loss": 0.0328,
+      "step": 2000
+    },
+    {
+      "epoch": 26.6,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 0.0335,
+      "step": 2500
+    },
+    {
+      "epoch": 31.91,
+      "learning_rate": 1.8085106382978724e-05,
+      "loss": 0.0616,
+      "step": 3000
+    },
+    {
+      "epoch": 37.23,
+      "learning_rate": 1.2765957446808511e-05,
+      "loss": 0.0913,
+      "step": 3500
+    },
+    {
+      "epoch": 42.55,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 0.182,
+      "step": 4000
+    },
+    {
+      "epoch": 47.87,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 0.5964,
+      "step": 4500
+    },
+    {
+      "epoch": 50.0,
+      "step": 4700,
+      "total_flos": 2484427852800000.0,
+      "train_loss": 0.2573057787469093,
+      "train_runtime": 1080.3341,
+      "train_samples_per_second": 138.846,
+      "train_steps_per_second": 4.351
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1880,
-  "num_train_epochs": 20,
+  "max_steps": 4700,
+  "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 993771141120000.0,
+  "total_flos": 2484427852800000.0,
   "trial_name": null,
   "trial_params": null
 }