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: 0.20056583736129271
+      value: 0.08150723019056925
 ---
 
 <!-- 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_stsb
 
-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 STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.3885
-- Pearson: 0.1998
-- Spearmanr: 0.2006
-- Combined Score: 0.2002
+- Loss: 2.3024
+- Pearson: 0.0671
+- Spearmanr: 0.0815
+- Combined Score: 0.0743
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 6.0,
+    "eval_combined_score": 0.07431917071856006,
+    "eval_loss": 2.3024191856384277,
+    "eval_pearson": 0.06713111124655088,
+    "eval_runtime": 2.4369,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 615.529,
+    "eval_spearmanr": 0.08150723019056925,
+    "eval_steps_per_second": 4.924,
+    "train_loss": 2.3151992232711227,
+    "train_runtime": 277.2499,
+    "train_samples": 5749,
+    "train_samples_per_second": 1036.79,
+    "train_steps_per_second": 8.115
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 6.0,
+    "eval_combined_score": 0.07431917071856006,
+    "eval_loss": 2.3024191856384277,
+    "eval_pearson": 0.06713111124655088,
+    "eval_runtime": 2.4369,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 615.529,
+    "eval_spearmanr": 0.08150723019056925,
+    "eval_steps_per_second": 4.924
+}

logs/events.out.tfevents.1674651429.garda.1933772.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6208cf535e0b05ebbcc00a200c6bd744c2567a59093ff240d8368385a93bcbc
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 2.3151992232711227,
+    "train_runtime": 277.2499,
+    "train_samples": 5749,
+    "train_samples_per_second": 1036.79,
+    "train_steps_per_second": 8.115
+}

trainer_state.json

@@ -0,0 +1,127 @@
+{
+  "best_metric": 2.3024191856384277,
+  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_stsb/checkpoint-45",
+  "epoch": 6.0,
+  "global_step": 270,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.906666666666667e-05,
+      "loss": 4.0455,
+      "step": 45
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.07431917071856006,
+      "eval_loss": 2.3024191856384277,
+      "eval_pearson": 0.06713111124655088,
+      "eval_runtime": 2.4381,
+      "eval_samples_per_second": 615.238,
+      "eval_spearmanr": 0.08150723019056925,
+      "eval_steps_per_second": 4.922,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.806666666666667e-05,
+      "loss": 2.1712,
+      "step": 90
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.06681067529441857,
+      "eval_loss": 2.6643781661987305,
+      "eval_pearson": 0.0611865012551173,
+      "eval_runtime": 2.4348,
+      "eval_samples_per_second": 616.068,
+      "eval_spearmanr": 0.07243484933371984,
+      "eval_steps_per_second": 4.929,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.706666666666667e-05,
+      "loss": 2.0637,
+      "step": 135
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.06251030424678386,
+      "eval_loss": 2.3624558448791504,
+      "eval_pearson": 0.058165468295460696,
+      "eval_runtime": 2.4296,
+      "eval_samples_per_second": 617.389,
+      "eval_spearmanr": 0.06685514019810702,
+      "eval_steps_per_second": 4.939,
+      "step": 135
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.606666666666667e-05,
+      "loss": 1.996,
+      "step": 180
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.07203733674672383,
+      "eval_loss": 2.8670907020568848,
+      "eval_pearson": 0.0712976536645173,
+      "eval_runtime": 2.4406,
+      "eval_samples_per_second": 614.594,
+      "eval_spearmanr": 0.07277701982893034,
+      "eval_steps_per_second": 4.917,
+      "step": 180
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5066666666666667e-05,
+      "loss": 1.908,
+      "step": 225
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.09261579480394588,
+      "eval_loss": 2.662240982055664,
+      "eval_pearson": 0.09544442363952416,
+      "eval_runtime": 2.4348,
+      "eval_samples_per_second": 616.07,
+      "eval_spearmanr": 0.08978716596836762,
+      "eval_steps_per_second": 4.929,
+      "step": 225
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.406666666666667e-05,
+      "loss": 1.7068,
+      "step": 270
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.2001877685151721,
+      "eval_loss": 2.388475179672241,
+      "eval_pearson": 0.19980969966905146,
+      "eval_runtime": 2.4335,
+      "eval_samples_per_second": 616.4,
+      "eval_spearmanr": 0.20056583736129271,
+      "eval_steps_per_second": 4.931,
+      "step": 270
+    },
+    {
+      "epoch": 6.0,
+      "step": 270,
+      "total_flos": 1081506722217984.0,
+      "train_loss": 2.3151992232711227,
+      "train_runtime": 277.2499,
+      "train_samples_per_second": 1036.79,
+      "train_steps_per_second": 8.115
+    }
+  ],
+  "max_steps": 2250,
+  "num_train_epochs": 50,
+  "total_flos": 1081506722217984.0,
+  "trial_name": null,
+  "trial_params": null
+}