End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 license: apache-2.0
 tags:
 - generated_from_trainer
@@ -14,7 +16,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MRPC
       type: glue
       config: mrpc
       split: validation
@@ -22,10 +24,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.6862745098039216
+      value: 0.6740196078431373
     - name: F1
       type: f1
-      value: 0.7664233576642336
+      value: 0.7787021630615641
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,12 +35,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mobilebert_sa_GLUE_Experiment_logit_kd_mrpc_128
 
-This model is a fine-tuned version of [google/mobilebert-uncased](https://huggingface.co/google/mobilebert-uncased) on the glue dataset.
+This model is a fine-tuned version of [google/mobilebert-uncased](https://huggingface.co/google/mobilebert-uncased) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5966
-- Accuracy: 0.6863
-- F1: 0.7664
-- Combined Score: 0.7263
+- Loss: 0.5213
+- Accuracy: 0.6740
+- F1: 0.7787
+- Combined Score: 0.7264
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.6740196078431373,
+    "eval_combined_score": 0.7263608854523507,
+    "eval_f1": 0.7787021630615641,
+    "eval_loss": 0.5212615728378296,
+    "eval_runtime": 0.7676,
+    "eval_samples": 408,
+    "eval_samples_per_second": 531.549,
+    "eval_steps_per_second": 5.211,
+    "train_loss": 0.47229831794212607,
+    "train_runtime": 291.7324,
+    "train_samples": 3668,
+    "train_samples_per_second": 628.658,
+    "train_steps_per_second": 4.97
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.6740196078431373,
+    "eval_combined_score": 0.7263608854523507,
+    "eval_f1": 0.7787021630615641,
+    "eval_loss": 0.5212615728378296,
+    "eval_runtime": 0.7676,
+    "eval_samples": 408,
+    "eval_samples_per_second": 531.549,
+    "eval_steps_per_second": 5.211
+}

logs/events.out.tfevents.1675030368.serv-3334.1021177.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b02bcedecc1ceceaaf7dbae37211d9b81e0b89950e613e767abdc08557160007
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 14.0,
+    "train_loss": 0.47229831794212607,
+    "train_runtime": 291.7324,
+    "train_samples": 3668,
+    "train_samples_per_second": 628.658,
+    "train_steps_per_second": 4.97
+}

trainer_state.json

@@ -0,0 +1,263 @@
+{
+  "best_metric": 0.5212615728378296,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_mrpc_128/checkpoint-261",
+  "epoch": 14.0,
+  "global_step": 406,
+  "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": 0.6368,
+      "step": 29
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.5564230680465698,
+      "eval_runtime": 0.7436,
+      "eval_samples_per_second": 548.712,
+      "eval_steps_per_second": 5.38,
+      "step": 29
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6099,
+      "step": 58
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.5557259321212769,
+      "eval_runtime": 0.746,
+      "eval_samples_per_second": 546.918,
+      "eval_steps_per_second": 5.362,
+      "step": 58
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.611,
+      "step": 87
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.5555070042610168,
+      "eval_runtime": 0.7446,
+      "eval_samples_per_second": 547.974,
+      "eval_steps_per_second": 5.372,
+      "step": 87
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.6101,
+      "step": 116
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.5567864179611206,
+      "eval_runtime": 0.7424,
+      "eval_samples_per_second": 549.533,
+      "eval_steps_per_second": 5.388,
+      "step": 116
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.608,
+      "step": 145
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.5539779663085938,
+      "eval_runtime": 0.7425,
+      "eval_samples_per_second": 549.496,
+      "eval_steps_per_second": 5.387,
+      "step": 145
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.6037,
+      "step": 174
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.5492002964019775,
+      "eval_runtime": 0.7401,
+      "eval_samples_per_second": 551.299,
+      "eval_steps_per_second": 5.405,
+      "step": 174
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.5761,
+      "step": 203
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6102941176470589,
+      "eval_combined_score": 0.647721316249272,
+      "eval_f1": 0.6851485148514851,
+      "eval_loss": 0.6065270900726318,
+      "eval_runtime": 0.7427,
+      "eval_samples_per_second": 549.382,
+      "eval_steps_per_second": 5.386,
+      "step": 203
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.4782,
+      "step": 232
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7331716191631292,
+      "eval_f1": 0.7800687285223368,
+      "eval_loss": 0.5340890884399414,
+      "eval_runtime": 0.7448,
+      "eval_samples_per_second": 547.827,
+      "eval_steps_per_second": 5.371,
+      "step": 232
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.4111,
+      "step": 261
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6740196078431373,
+      "eval_combined_score": 0.7263608854523507,
+      "eval_f1": 0.7787021630615641,
+      "eval_loss": 0.5212615728378296,
+      "eval_runtime": 0.7427,
+      "eval_samples_per_second": 549.311,
+      "eval_steps_per_second": 5.385,
+      "step": 261
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.3526,
+      "step": 290
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.736470588235294,
+      "eval_f1": 0.7866666666666665,
+      "eval_loss": 0.57920241355896,
+      "eval_runtime": 0.7437,
+      "eval_samples_per_second": 548.588,
+      "eval_steps_per_second": 5.378,
+      "step": 290
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.3188,
+      "step": 319
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7350069276368867,
+      "eval_f1": 0.7763864042933811,
+      "eval_loss": 0.5759749412536621,
+      "eval_runtime": 0.7403,
+      "eval_samples_per_second": 551.091,
+      "eval_steps_per_second": 5.403,
+      "step": 319
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.2918,
+      "step": 348
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.6911764705882353,
+      "eval_combined_score": 0.7395276292335116,
+      "eval_f1": 0.787878787878788,
+      "eval_loss": 0.6406097412109375,
+      "eval_runtime": 0.7431,
+      "eval_samples_per_second": 549.016,
+      "eval_steps_per_second": 5.383,
+      "step": 348
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7e-05,
+      "loss": 0.2568,
+      "step": 377
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.6764705882352942,
+      "eval_combined_score": 0.715100965759438,
+      "eval_f1": 0.753731343283582,
+      "eval_loss": 0.5908408761024475,
+      "eval_runtime": 0.7434,
+      "eval_samples_per_second": 548.842,
+      "eval_steps_per_second": 5.381,
+      "step": 377
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.2472,
+      "step": 406
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7263489337340776,
+      "eval_f1": 0.7664233576642336,
+      "eval_loss": 0.5965602993965149,
+      "eval_runtime": 0.7403,
+      "eval_samples_per_second": 551.135,
+      "eval_steps_per_second": 5.403,
+      "step": 406
+    },
+    {
+      "epoch": 14.0,
+      "step": 406,
+      "total_flos": 1223954094096384.0,
+      "train_loss": 0.47229831794212607,
+      "train_runtime": 291.7324,
+      "train_samples_per_second": 628.658,
+      "train_steps_per_second": 4.97
+    }
+  ],
+  "max_steps": 1450,
+  "num_train_epochs": 50,
+  "total_flos": 1223954094096384.0,
+  "trial_name": null,
+  "trial_params": null
+}