End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MRPC
       type: glue
       config: mrpc
       split: validation
@@ -21,10 +23,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.6593137254901961
+      value: 0.7083333333333334
     - name: F1
       type: f1
-      value: 0.7599309153713298
+      value: 0.8172043010752688
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,12 +34,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_new_pretrain_48_emb_com_mrpc
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_emb_compress_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_emb_compress_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_emb_compress_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_emb_compress_48) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9040
-- Accuracy: 0.6593
-- F1: 0.7599
-- Combined Score: 0.7096
+- Loss: 0.5831
+- Accuracy: 0.7083
+- F1: 0.8172
+- Combined Score: 0.7628
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.7083333333333334,
+    "eval_combined_score": 0.762768817204301,
+    "eval_f1": 0.8172043010752688,
+    "eval_loss": 0.5831031799316406,
+    "eval_runtime": 1.8429,
+    "eval_samples": 408,
+    "eval_samples_per_second": 221.39,
+    "eval_steps_per_second": 2.17,
+    "train_loss": 0.5139258853320418,
+    "train_runtime": 501.0101,
+    "train_samples": 3668,
+    "train_samples_per_second": 366.06,
+    "train_steps_per_second": 2.894
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.7083333333333334,
+    "eval_combined_score": 0.762768817204301,
+    "eval_f1": 0.8172043010752688,
+    "eval_loss": 0.5831031799316406,
+    "eval_runtime": 1.8429,
+    "eval_samples": 408,
+    "eval_samples_per_second": 221.39,
+    "eval_steps_per_second": 2.17
+}

logs/events.out.tfevents.1686759393.garda.2601621.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c697ffe227d7410b6fd674ca38340e9902cbdff7462fbe71042b4a8a40c401f8
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.5139258853320418,
+    "train_runtime": 501.0101,
+    "train_samples": 3668,
+    "train_samples_per_second": 366.06,
+    "train_steps_per_second": 2.894
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_metric": 0.5831031799316406,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_48_emb_com_mrpc/checkpoint-87",
+  "epoch": 8.0,
+  "global_step": 232,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.6776,
+      "step": 29
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.7577850289051791,
+      "eval_f1": 0.8145896656534954,
+      "eval_loss": 0.6051964163780212,
+      "eval_runtime": 1.8582,
+      "eval_samples_per_second": 219.568,
+      "eval_steps_per_second": 2.153,
+      "step": 29
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6271,
+      "step": 58
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.696078431372549,
+      "eval_combined_score": 0.749313101036593,
+      "eval_f1": 0.8025477707006369,
+      "eval_loss": 0.6111815571784973,
+      "eval_runtime": 1.8617,
+      "eval_samples_per_second": 219.152,
+      "eval_steps_per_second": 2.149,
+      "step": 58
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.58,
+      "step": 87
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7083333333333334,
+      "eval_combined_score": 0.762768817204301,
+      "eval_f1": 0.8172043010752688,
+      "eval_loss": 0.5831031799316406,
+      "eval_runtime": 1.8481,
+      "eval_samples_per_second": 220.766,
+      "eval_steps_per_second": 2.164,
+      "step": 87
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.5494,
+      "step": 116
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.7551776960784313,
+      "eval_f1": 0.809375,
+      "eval_loss": 0.6457833051681519,
+      "eval_runtime": 1.8552,
+      "eval_samples_per_second": 219.925,
+      "eval_steps_per_second": 2.156,
+      "step": 116
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.5148,
+      "step": 145
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7360004346566213,
+      "eval_f1": 0.7881773399014779,
+      "eval_loss": 0.6067324280738831,
+      "eval_runtime": 1.8513,
+      "eval_samples_per_second": 220.386,
+      "eval_steps_per_second": 2.161,
+      "step": 145
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.4573,
+      "step": 174
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.739911448450348,
+      "eval_f1": 0.7935483870967743,
+      "eval_loss": 0.6266855597496033,
+      "eval_runtime": 1.8536,
+      "eval_samples_per_second": 220.109,
+      "eval_steps_per_second": 2.158,
+      "step": 174
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.395,
+      "step": 203
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6274509803921569,
+      "eval_combined_score": 0.6784941734700998,
+      "eval_f1": 0.7295373665480426,
+      "eval_loss": 0.789704442024231,
+      "eval_runtime": 1.8594,
+      "eval_samples_per_second": 219.425,
+      "eval_steps_per_second": 2.151,
+      "step": 203
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.3102,
+      "step": 232
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6593137254901961,
+      "eval_combined_score": 0.709622320430763,
+      "eval_f1": 0.7599309153713298,
+      "eval_loss": 0.9040012955665588,
+      "eval_runtime": 1.8564,
+      "eval_samples_per_second": 219.778,
+      "eval_steps_per_second": 2.155,
+      "step": 232
+    },
+    {
+      "epoch": 8.0,
+      "step": 232,
+      "total_flos": 4350159504801792.0,
+      "train_loss": 0.5139258853320418,
+      "train_runtime": 501.0101,
+      "train_samples_per_second": 366.06,
+      "train_steps_per_second": 2.894
+    }
+  ],
+  "max_steps": 1450,
+  "num_train_epochs": 50,
+  "total_flos": 4350159504801792.0,
+  "trial_name": null,
+  "trial_params": null
+}