End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -12,7 +14,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.3273560876209883
+      value: 0.318246541903987
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_new_pretrain_w_init__mnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0990
-- Accuracy: 0.3274
+- Loss: 1.0963
+- Accuracy: 0.3182
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 7.0,
+    "epoch_mm": 7.0,
+    "eval_accuracy": 0.31818644931227713,
+    "eval_accuracy_mm": 0.318246541903987,
+    "eval_loss": 1.0963075160980225,
+    "eval_loss_mm": 1.0963244438171387,
+    "eval_runtime": 17.5761,
+    "eval_runtime_mm": 17.6104,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 558.429,
+    "eval_samples_per_second_mm": 558.305,
+    "eval_steps_per_second": 4.381,
+    "eval_steps_per_second_mm": 4.372,
+    "train_loss": 1.1130940406587586,
+    "train_runtime": 14060.0313,
+    "train_samples": 392702,
+    "train_samples_per_second": 1396.519,
+    "train_steps_per_second": 10.91
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 7.0,
+    "epoch_mm": 7.0,
+    "eval_accuracy": 0.31818644931227713,
+    "eval_accuracy_mm": 0.318246541903987,
+    "eval_loss": 1.0963075160980225,
+    "eval_loss_mm": 1.0963244438171387,
+    "eval_runtime": 17.5761,
+    "eval_runtime_mm": 17.6104,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 558.429,
+    "eval_samples_per_second_mm": 558.305,
+    "eval_steps_per_second": 4.381,
+    "eval_steps_per_second_mm": 4.372
+}

logs/events.out.tfevents.1685566753.serv-9223.3032138.26

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4155058d96b6b84dd55386faed61c079e2ca33aba65287193c0fef024d5e5957
+size 698

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 1.1130940406587586,
+    "train_runtime": 14060.0313,
+    "train_samples": 392702,
+    "train_samples_per_second": 1396.519,
+    "train_steps_per_second": 10.91
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 1.0963075160980225,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__mnli/checkpoint-6136",
+  "epoch": 7.0,
+  "global_step": 21476,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00049,
+      "loss": 1.1346,
+      "step": 3068
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0987932682037354,
+      "eval_runtime": 17.7369,
+      "eval_samples_per_second": 553.367,
+      "eval_steps_per_second": 4.341,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048,
+      "loss": 1.1013,
+      "step": 6136
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0963075160980225,
+      "eval_runtime": 17.7358,
+      "eval_samples_per_second": 553.402,
+      "eval_steps_per_second": 4.342,
+      "step": 6136
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047,
+      "loss": 1.1611,
+      "step": 9204
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.0990049839019775,
+      "eval_runtime": 17.7014,
+      "eval_samples_per_second": 554.477,
+      "eval_steps_per_second": 4.35,
+      "step": 9204
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046,
+      "loss": 1.0988,
+      "step": 12272
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0987932682037354,
+      "eval_runtime": 17.7275,
+      "eval_samples_per_second": 553.658,
+      "eval_steps_per_second": 4.344,
+      "step": 12272
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 1.0987,
+      "step": 15340
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0987932682037354,
+      "eval_runtime": 17.7181,
+      "eval_samples_per_second": 553.953,
+      "eval_steps_per_second": 4.346,
+      "step": 15340
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044,
+      "loss": 1.0985,
+      "step": 18408
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.0990049839019775,
+      "eval_runtime": 17.7077,
+      "eval_samples_per_second": 554.278,
+      "eval_steps_per_second": 4.348,
+      "step": 18408
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00043,
+      "loss": 1.0987,
+      "step": 21476
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.0990049839019775,
+      "eval_runtime": 17.6862,
+      "eval_samples_per_second": 554.953,
+      "eval_steps_per_second": 4.354,
+      "step": 21476
+    },
+    {
+      "epoch": 7.0,
+      "step": 21476,
+      "total_flos": 4.071085382597345e+17,
+      "train_loss": 1.1130940406587586,
+      "train_runtime": 14060.0313,
+      "train_samples_per_second": 1396.519,
+      "train_steps_per_second": 10.91
+    }
+  ],
+  "max_steps": 153400,
+  "num_train_epochs": 50,
+  "total_flos": 4.071085382597345e+17,
+  "trial_name": null,
+  "trial_params": null
+}