End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 license: apache-2.0
 tags:
 - generated_from_trainer
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE COLA
       type: glue
       config: cola
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Matthews Correlation
       type: matthews_correlation
-      value: 0.5422671648885407
+      value: 0.5439723028804963
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_cola
 
-This model is a fine-tuned version of [gokuls/mobilebert_sa_pre-training-complete](https://huggingface.co/gokuls/mobilebert_sa_pre-training-complete) on the glue dataset.
+This model is a fine-tuned version of [gokuls/mobilebert_sa_pre-training-complete](https://huggingface.co/gokuls/mobilebert_sa_pre-training-complete) on the GLUE COLA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4125
-- Matthews Correlation: 0.5423
+- Loss: 0.3784
+- Matthews Correlation: 0.5440
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 9.0,
+    "eval_loss": 0.3783751130104065,
+    "eval_matthews_correlation": 0.5439723028804963,
+    "eval_runtime": 1.989,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 524.378,
+    "eval_steps_per_second": 4.525,
+    "train_loss": 0.23667547200647357,
+    "train_runtime": 438.7548,
+    "train_samples": 8551,
+    "train_samples_per_second": 974.462,
+    "train_steps_per_second": 7.635
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "eval_loss": 0.3783751130104065,
+    "eval_matthews_correlation": 0.5439723028804963,
+    "eval_runtime": 1.989,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 524.378,
+    "eval_steps_per_second": 4.525
+}

logs/events.out.tfevents.1675031366.serv-3334.1119753.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c94aa668cddd667c64bcc50c4f9653e830f0c73da5ca45e8a77cfac070eec0fd
+size 375

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 0.23667547200647357,
+    "train_runtime": 438.7548,
+    "train_samples": 8551,
+    "train_samples_per_second": 974.462,
+    "train_steps_per_second": 7.635
+}

trainer_state.json

@@ -0,0 +1,160 @@
+{
+  "best_metric": 0.3783751130104065,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_cola/checkpoint-268",
+  "epoch": 9.0,
+  "global_step": 603,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6286,
+      "step": 67
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.4351368844509125,
+      "eval_matthews_correlation": 0.4468514158987457,
+      "eval_runtime": 1.9774,
+      "eval_samples_per_second": 527.47,
+      "eval_steps_per_second": 4.552,
+      "step": 67
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3982,
+      "step": 134
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.40038201212882996,
+      "eval_matthews_correlation": 0.48584699342231225,
+      "eval_runtime": 1.9801,
+      "eval_samples_per_second": 526.742,
+      "eval_steps_per_second": 4.545,
+      "step": 134
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.2722,
+      "step": 201
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.38673949241638184,
+      "eval_matthews_correlation": 0.5371884165342511,
+      "eval_runtime": 1.9718,
+      "eval_samples_per_second": 528.945,
+      "eval_steps_per_second": 4.564,
+      "step": 201
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2056,
+      "step": 268
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.3783751130104065,
+      "eval_matthews_correlation": 0.5439723028804963,
+      "eval_runtime": 1.9764,
+      "eval_samples_per_second": 527.735,
+      "eval_steps_per_second": 4.554,
+      "step": 268
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1649,
+      "step": 335
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.4273626506328583,
+      "eval_matthews_correlation": 0.562419829892001,
+      "eval_runtime": 1.9713,
+      "eval_samples_per_second": 529.083,
+      "eval_steps_per_second": 4.565,
+      "step": 335
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.135,
+      "step": 402
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.44399121403694153,
+      "eval_matthews_correlation": 0.5261360919387726,
+      "eval_runtime": 1.9804,
+      "eval_samples_per_second": 526.651,
+      "eval_steps_per_second": 4.544,
+      "step": 402
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.1174,
+      "step": 469
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.45434868335723877,
+      "eval_matthews_correlation": 0.5494701108221334,
+      "eval_runtime": 1.9695,
+      "eval_samples_per_second": 529.58,
+      "eval_steps_per_second": 4.57,
+      "step": 469
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.108,
+      "step": 536
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.3884676992893219,
+      "eval_matthews_correlation": 0.550612081256011,
+      "eval_runtime": 1.9709,
+      "eval_samples_per_second": 529.205,
+      "eval_steps_per_second": 4.566,
+      "step": 536
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.1002,
+      "step": 603
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 0.41253650188446045,
+      "eval_matthews_correlation": 0.5422671648885407,
+      "eval_runtime": 1.9646,
+      "eval_samples_per_second": 530.903,
+      "eval_steps_per_second": 4.581,
+      "step": 603
+    },
+    {
+      "epoch": 9.0,
+      "step": 603,
+      "total_flos": 2444040855355392.0,
+      "train_loss": 0.23667547200647357,
+      "train_runtime": 438.7548,
+      "train_samples_per_second": 974.462,
+      "train_steps_per_second": 7.635
+    }
+  ],
+  "max_steps": 3350,
+  "num_train_epochs": 50,
+  "total_flos": 2444040855355392.0,
+  "trial_name": null,
+  "trial_params": null
+}