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 STSB
       type: glue
       config: stsb
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Spearmanr
       type: spearmanr
-      value: 0.8624005783710303
+      value: 0.8642221596976783
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,12 +31,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_stsb
 
-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 STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2985
-- Pearson: 0.8647
-- Spearmanr: 0.8624
-- Combined Score: 0.8636
+- Loss: 0.2919
+- Pearson: 0.8665
+- Spearmanr: 0.8642
+- Combined Score: 0.8654
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 20.0,
+    "eval_combined_score": 0.8653611315954421,
+    "eval_loss": 0.2918669879436493,
+    "eval_pearson": 0.8665001034932058,
+    "eval_runtime": 2.8822,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 520.444,
+    "eval_spearmanr": 0.8642221596976783,
+    "eval_steps_per_second": 4.164,
+    "train_loss": 0.14503458228376176,
+    "train_runtime": 766.7055,
+    "train_samples": 5749,
+    "train_samples_per_second": 374.916,
+    "train_steps_per_second": 2.935
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 20.0,
+    "eval_combined_score": 0.8653611315954421,
+    "eval_loss": 0.2918669879436493,
+    "eval_pearson": 0.8665001034932058,
+    "eval_runtime": 2.8822,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 520.444,
+    "eval_spearmanr": 0.8642221596976783,
+    "eval_steps_per_second": 4.164
+}

logs/events.out.tfevents.1675096312.serv-3334.1119753.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4663696415cb9b0e3e777dfb63b8518ec7d4858c7eba945d3ef7641cf318770
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "train_loss": 0.14503458228376176,
+    "train_runtime": 766.7055,
+    "train_samples": 5749,
+    "train_samples_per_second": 374.916,
+    "train_steps_per_second": 2.935
+}

trainer_state.json

@@ -0,0 +1,365 @@
+{
+  "best_metric": 0.2918669879436493,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_stsb/checkpoint-675",
+  "epoch": 20.0,
+  "global_step": 900,
+  "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": 1.1501,
+      "step": 45
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.7848155232626217,
+      "eval_loss": 0.47263744473457336,
+      "eval_pearson": 0.7774458991104952,
+      "eval_runtime": 2.8147,
+      "eval_samples_per_second": 532.917,
+      "eval_spearmanr": 0.7921851474147482,
+      "eval_steps_per_second": 4.263,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.364,
+      "step": 90
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.8455802664916386,
+      "eval_loss": 0.34795352816581726,
+      "eval_pearson": 0.8456719627096524,
+      "eval_runtime": 2.8104,
+      "eval_samples_per_second": 533.726,
+      "eval_spearmanr": 0.8454885702736249,
+      "eval_steps_per_second": 4.27,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.259,
+      "step": 135
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.8585617866351674,
+      "eval_loss": 0.3156156539916992,
+      "eval_pearson": 0.858172801547449,
+      "eval_runtime": 2.841,
+      "eval_samples_per_second": 527.982,
+      "eval_spearmanr": 0.8589507717228858,
+      "eval_steps_per_second": 4.224,
+      "step": 135
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2054,
+      "step": 180
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.8549704899730026,
+      "eval_loss": 0.423094242811203,
+      "eval_pearson": 0.8550544334442006,
+      "eval_runtime": 2.8159,
+      "eval_samples_per_second": 532.695,
+      "eval_spearmanr": 0.8548865465018045,
+      "eval_steps_per_second": 4.262,
+      "step": 180
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1629,
+      "step": 225
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.8661052960846953,
+      "eval_loss": 0.3244686424732208,
+      "eval_pearson": 0.8668446586721916,
+      "eval_runtime": 2.8146,
+      "eval_samples_per_second": 532.932,
+      "eval_spearmanr": 0.8653659334971991,
+      "eval_steps_per_second": 4.263,
+      "step": 225
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1263,
+      "step": 270
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.8637024825465704,
+      "eval_loss": 0.31924131512641907,
+      "eval_pearson": 0.8648820058548446,
+      "eval_runtime": 2.8102,
+      "eval_samples_per_second": 533.763,
+      "eval_spearmanr": 0.8625229592382962,
+      "eval_steps_per_second": 4.27,
+      "step": 270
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.1021,
+      "step": 315
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.8641855325700929,
+      "eval_loss": 0.33367273211479187,
+      "eval_pearson": 0.8655048921422074,
+      "eval_runtime": 2.824,
+      "eval_samples_per_second": 531.159,
+      "eval_spearmanr": 0.8628661729979783,
+      "eval_steps_per_second": 4.249,
+      "step": 315
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.0841,
+      "step": 360
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.8589275349426662,
+      "eval_loss": 0.3060952126979828,
+      "eval_pearson": 0.8601387509705535,
+      "eval_runtime": 2.8107,
+      "eval_samples_per_second": 533.667,
+      "eval_spearmanr": 0.8577163189147787,
+      "eval_steps_per_second": 4.269,
+      "step": 360
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.0713,
+      "step": 405
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.8565713298409787,
+      "eval_loss": 0.36002376675605774,
+      "eval_pearson": 0.8576436109786436,
+      "eval_runtime": 2.8144,
+      "eval_samples_per_second": 532.978,
+      "eval_spearmanr": 0.8554990487033137,
+      "eval_steps_per_second": 4.264,
+      "step": 405
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.0587,
+      "step": 450
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.8609998923935287,
+      "eval_loss": 0.31353628635406494,
+      "eval_pearson": 0.8619690194372541,
+      "eval_runtime": 2.8144,
+      "eval_samples_per_second": 532.966,
+      "eval_spearmanr": 0.8600307653498033,
+      "eval_steps_per_second": 4.264,
+      "step": 450
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.0488,
+      "step": 495
+    },
+    {
+      "epoch": 11.0,
+      "eval_combined_score": 0.8630528258320453,
+      "eval_loss": 0.3006461262702942,
+      "eval_pearson": 0.8640987141337612,
+      "eval_runtime": 2.8106,
+      "eval_samples_per_second": 533.7,
+      "eval_spearmanr": 0.8620069375303294,
+      "eval_steps_per_second": 4.27,
+      "step": 495
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.0441,
+      "step": 540
+    },
+    {
+      "epoch": 12.0,
+      "eval_combined_score": 0.8633072379827798,
+      "eval_loss": 0.3308115303516388,
+      "eval_pearson": 0.8645046486846006,
+      "eval_runtime": 2.8091,
+      "eval_samples_per_second": 533.981,
+      "eval_spearmanr": 0.8621098272809588,
+      "eval_steps_per_second": 4.272,
+      "step": 540
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7e-05,
+      "loss": 0.0385,
+      "step": 585
+    },
+    {
+      "epoch": 13.0,
+      "eval_combined_score": 0.8610427119151247,
+      "eval_loss": 0.34676018357276917,
+      "eval_pearson": 0.8619973007896685,
+      "eval_runtime": 2.8063,
+      "eval_samples_per_second": 534.504,
+      "eval_spearmanr": 0.860088123040581,
+      "eval_steps_per_second": 4.276,
+      "step": 585
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.0346,
+      "step": 630
+    },
+    {
+      "epoch": 14.0,
+      "eval_combined_score": 0.864625467802823,
+      "eval_loss": 0.3174898028373718,
+      "eval_pearson": 0.8658273441962089,
+      "eval_runtime": 2.8116,
+      "eval_samples_per_second": 533.507,
+      "eval_spearmanr": 0.8634235914094371,
+      "eval_steps_per_second": 4.268,
+      "step": 630
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 3.5e-05,
+      "loss": 0.0298,
+      "step": 675
+    },
+    {
+      "epoch": 15.0,
+      "eval_combined_score": 0.8653611315954421,
+      "eval_loss": 0.2918669879436493,
+      "eval_pearson": 0.8665001034932058,
+      "eval_runtime": 2.812,
+      "eval_samples_per_second": 533.434,
+      "eval_spearmanr": 0.8642221596976783,
+      "eval_steps_per_second": 4.267,
+      "step": 675
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.0299,
+      "step": 720
+    },
+    {
+      "epoch": 16.0,
+      "eval_combined_score": 0.8638818647743551,
+      "eval_loss": 0.310304194688797,
+      "eval_pearson": 0.8649166348556964,
+      "eval_runtime": 2.8123,
+      "eval_samples_per_second": 533.376,
+      "eval_spearmanr": 0.862847094693014,
+      "eval_steps_per_second": 4.267,
+      "step": 720
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 3.3e-05,
+      "loss": 0.0263,
+      "step": 765
+    },
+    {
+      "epoch": 17.0,
+      "eval_combined_score": 0.8609287949422196,
+      "eval_loss": 0.3324874937534332,
+      "eval_pearson": 0.8619612623497133,
+      "eval_runtime": 2.8136,
+      "eval_samples_per_second": 533.131,
+      "eval_spearmanr": 0.859896327534726,
+      "eval_steps_per_second": 4.265,
+      "step": 765
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.0237,
+      "step": 810
+    },
+    {
+      "epoch": 18.0,
+      "eval_combined_score": 0.8623362746505621,
+      "eval_loss": 0.3092164397239685,
+      "eval_pearson": 0.8635793188062354,
+      "eval_runtime": 2.8105,
+      "eval_samples_per_second": 533.72,
+      "eval_spearmanr": 0.8610932304948888,
+      "eval_steps_per_second": 4.27,
+      "step": 810
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 3.1e-05,
+      "loss": 0.0213,
+      "step": 855
+    },
+    {
+      "epoch": 19.0,
+      "eval_combined_score": 0.8641800158406565,
+      "eval_loss": 0.31693407893180847,
+      "eval_pearson": 0.8652586147489575,
+      "eval_runtime": 2.8089,
+      "eval_samples_per_second": 534.009,
+      "eval_spearmanr": 0.8631014169323554,
+      "eval_steps_per_second": 4.272,
+      "step": 855
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 3e-05,
+      "loss": 0.0196,
+      "step": 900
+    },
+    {
+      "epoch": 20.0,
+      "eval_combined_score": 0.8635686954927306,
+      "eval_loss": 0.29848989844322205,
+      "eval_pearson": 0.8647368126144308,
+      "eval_runtime": 2.8089,
+      "eval_samples_per_second": 534.023,
+      "eval_spearmanr": 0.8624005783710303,
+      "eval_steps_per_second": 4.272,
+      "step": 900
+    },
+    {
+      "epoch": 20.0,
+      "step": 900,
+      "total_flos": 3651409731911680.0,
+      "train_loss": 0.14503458228376176,
+      "train_runtime": 766.7055,
+      "train_samples_per_second": 374.916,
+      "train_steps_per_second": 2.935
+    }
+  ],
+  "max_steps": 2250,
+  "num_train_epochs": 50,
+  "total_flos": 3651409731911680.0,
+  "trial_name": null,
+  "trial_params": null
+}