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 QQP
       type: glue
       config: qqp
       split: validation
@@ -21,10 +23,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.855107593371259
+      value: 0.8430373485035865
     - name: F1
       type: f1
-      value: 0.8099409512685743
+      value: 0.7845307619176966
 ---
 
 <!-- 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_w_init_48_qqp
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3726
-- Accuracy: 0.8551
-- F1: 0.8099
-- Combined Score: 0.8325
+- Loss: 0.3476
+- Accuracy: 0.8430
+- F1: 0.7845
+- Combined Score: 0.8138
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8430373485035865,
+    "eval_combined_score": 0.8137840552106415,
+    "eval_f1": 0.7845307619176966,
+    "eval_loss": 0.3475593626499176,
+    "eval_runtime": 74.9788,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 539.219,
+    "eval_steps_per_second": 4.215,
+    "train_loss": 0.2936604352555894,
+    "train_runtime": 15852.884,
+    "train_samples": 363846,
+    "train_samples_per_second": 1147.57,
+    "train_steps_per_second": 8.967
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8430373485035865,
+    "eval_combined_score": 0.8137840552106415,
+    "eval_f1": 0.7845307619176966,
+    "eval_loss": 0.3475593626499176,
+    "eval_runtime": 74.9788,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 539.219,
+    "eval_steps_per_second": 4.215
+}

logs/events.out.tfevents.1686057281.serv-9207.2389173.14

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d184902c261421f5ac1e8a8f113e3b6a38458a3eb0123aac9d4bf9bcce4128b
+size 475

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.2936604352555894,
+    "train_runtime": 15852.884,
+    "train_samples": 363846,
+    "train_samples_per_second": 1147.57,
+    "train_steps_per_second": 8.967
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_metric": 0.3475593626499176,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init_48_qqp/checkpoint-8529",
+  "epoch": 8.0,
+  "global_step": 22744,
+  "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": 0.4637,
+      "step": 2843
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8136284936928023,
+      "eval_combined_score": 0.7886255841742551,
+      "eval_f1": 0.763622674655708,
+      "eval_loss": 0.39073047041893005,
+      "eval_runtime": 74.9641,
+      "eval_samples_per_second": 539.325,
+      "eval_steps_per_second": 4.215,
+      "step": 2843
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.363,
+      "step": 5686
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8337620578778135,
+      "eval_combined_score": 0.8118949312788737,
+      "eval_f1": 0.7900278046799338,
+      "eval_loss": 0.35355764627456665,
+      "eval_runtime": 75.0974,
+      "eval_samples_per_second": 538.368,
+      "eval_steps_per_second": 4.208,
+      "step": 5686
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.3211,
+      "step": 8529
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8430373485035865,
+      "eval_combined_score": 0.8137840552106415,
+      "eval_f1": 0.7845307619176966,
+      "eval_loss": 0.3475593626499176,
+      "eval_runtime": 74.3052,
+      "eval_samples_per_second": 544.107,
+      "eval_steps_per_second": 4.253,
+      "step": 8529
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.2906,
+      "step": 11372
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.853104130596092,
+      "eval_combined_score": 0.8294810313338081,
+      "eval_f1": 0.8058579320715243,
+      "eval_loss": 0.35388582944869995,
+      "eval_runtime": 74.2817,
+      "eval_samples_per_second": 544.28,
+      "eval_steps_per_second": 4.254,
+      "step": 11372
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.2603,
+      "step": 14215
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8530546623794212,
+      "eval_combined_score": 0.8273883931661482,
+      "eval_f1": 0.8017221239528752,
+      "eval_loss": 0.3530649244785309,
+      "eval_runtime": 75.3657,
+      "eval_samples_per_second": 536.451,
+      "eval_steps_per_second": 4.193,
+      "step": 14215
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.2373,
+      "step": 17058
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8560722235963394,
+      "eval_combined_score": 0.8324891675270456,
+      "eval_f1": 0.8089061114577517,
+      "eval_loss": 0.3716297447681427,
+      "eval_runtime": 75.0931,
+      "eval_samples_per_second": 538.399,
+      "eval_steps_per_second": 4.208,
+      "step": 17058
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.2175,
+      "step": 19901
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8564679693297057,
+      "eval_combined_score": 0.8343737963945459,
+      "eval_f1": 0.812279623459386,
+      "eval_loss": 0.35529038310050964,
+      "eval_runtime": 74.8508,
+      "eval_samples_per_second": 540.141,
+      "eval_steps_per_second": 4.222,
+      "step": 19901
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.1957,
+      "step": 22744
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.855107593371259,
+      "eval_combined_score": 0.8325242723199167,
+      "eval_f1": 0.8099409512685743,
+      "eval_loss": 0.3725973963737488,
+      "eval_runtime": 75.7858,
+      "eval_samples_per_second": 533.477,
+      "eval_steps_per_second": 4.17,
+      "step": 22744
+    },
+    {
+      "epoch": 8.0,
+      "step": 22744,
+      "total_flos": 4.310753534195794e+17,
+      "train_loss": 0.2936604352555894,
+      "train_runtime": 15852.884,
+      "train_samples_per_second": 1147.57,
+      "train_steps_per_second": 8.967
+    }
+  ],
+  "max_steps": 142150,
+  "num_train_epochs": 50,
+  "total_flos": 4.310753534195794e+17,
+  "trial_name": null,
+  "trial_params": null
+}