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 SST2
       type: glue
       config: sst2
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7958715596330275
+      value: 0.7947247706422018
 ---
 
 <!-- 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_48_emb_com_sst2
 
-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 SST2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7695
-- Accuracy: 0.7959
+- Loss: 0.4656
+- Accuracy: 0.7947
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.7947247706422018,
+    "eval_loss": 0.4656488299369812,
+    "eval_runtime": 3.9144,
+    "eval_samples": 872,
+    "eval_samples_per_second": 222.766,
+    "eval_steps_per_second": 1.788,
+    "train_loss": 0.19084218511997636,
+    "train_runtime": 9385.9864,
+    "train_samples": 67349,
+    "train_samples_per_second": 358.774,
+    "train_steps_per_second": 2.807
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.7947247706422018,
+    "eval_loss": 0.4656488299369812,
+    "eval_runtime": 3.9144,
+    "eval_samples": 872,
+    "eval_samples_per_second": 222.766,
+    "eval_steps_per_second": 1.788
+}

logs/events.out.tfevents.1686757579.garda.2601621.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d43c83f2da6b8353c5e0b254bb792ce356d34d89e970eedc8cb6383432c129aa
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 0.19084218511997636,
+    "train_runtime": 9385.9864,
+    "train_samples": 67349,
+    "train_samples_per_second": 358.774,
+    "train_steps_per_second": 2.807
+}

trainer_state.json

@@ -0,0 +1,175 @@
+{
+  "best_metric": 0.4656488299369812,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_48_emb_com_sst2/checkpoint-2635",
+  "epoch": 10.0,
+  "global_step": 5270,
+  "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": 0.4711,
+      "step": 527
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7786697247706422,
+      "eval_loss": 0.47297555208206177,
+      "eval_runtime": 3.9484,
+      "eval_samples_per_second": 220.85,
+      "eval_steps_per_second": 1.773,
+      "step": 527
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.2942,
+      "step": 1054
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7993119266055045,
+      "eval_loss": 0.5416545867919922,
+      "eval_runtime": 3.9198,
+      "eval_samples_per_second": 222.46,
+      "eval_steps_per_second": 1.786,
+      "step": 1054
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.228,
+      "step": 1581
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7901376146788991,
+      "eval_loss": 0.5139726400375366,
+      "eval_runtime": 3.92,
+      "eval_samples_per_second": 222.446,
+      "eval_steps_per_second": 1.786,
+      "step": 1581
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.1897,
+      "step": 2108
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7947247706422018,
+      "eval_loss": 0.49514883756637573,
+      "eval_runtime": 3.925,
+      "eval_samples_per_second": 222.163,
+      "eval_steps_per_second": 1.783,
+      "step": 2108
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.1608,
+      "step": 2635
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7947247706422018,
+      "eval_loss": 0.4656488299369812,
+      "eval_runtime": 3.9285,
+      "eval_samples_per_second": 221.967,
+      "eval_steps_per_second": 1.782,
+      "step": 2635
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.1422,
+      "step": 3162
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7740825688073395,
+      "eval_loss": 0.6786444187164307,
+      "eval_runtime": 3.9166,
+      "eval_samples_per_second": 222.641,
+      "eval_steps_per_second": 1.787,
+      "step": 3162
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.1244,
+      "step": 3689
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8027522935779816,
+      "eval_loss": 0.6885600686073303,
+      "eval_runtime": 3.9243,
+      "eval_samples_per_second": 222.207,
+      "eval_steps_per_second": 1.784,
+      "step": 3689
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.1112,
+      "step": 4216
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7821100917431193,
+      "eval_loss": 0.7387205362319946,
+      "eval_runtime": 3.9137,
+      "eval_samples_per_second": 222.805,
+      "eval_steps_per_second": 1.789,
+      "step": 4216
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.0985,
+      "step": 4743
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7935779816513762,
+      "eval_loss": 0.7169041037559509,
+      "eval_runtime": 3.9186,
+      "eval_samples_per_second": 222.526,
+      "eval_steps_per_second": 1.786,
+      "step": 4743
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.0883,
+      "step": 5270
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7958715596330275,
+      "eval_loss": 0.7695210576057434,
+      "eval_runtime": 3.9484,
+      "eval_samples_per_second": 220.85,
+      "eval_steps_per_second": 1.773,
+      "step": 5270
+    },
+    {
+      "epoch": 10.0,
+      "step": 5270,
+      "total_flos": 9.984286060969984e+16,
+      "train_loss": 0.19084218511997636,
+      "train_runtime": 9385.9864,
+      "train_samples_per_second": 358.774,
+      "train_steps_per_second": 2.807
+    }
+  ],
+  "max_steps": 26350,
+  "num_train_epochs": 50,
+  "total_flos": 9.984286060969984e+16,
+  "trial_name": null,
+  "trial_params": null
+}