End of training

README.md

@@ -1,15 +1,33 @@
 ---
 library_name: transformers
+language:
+- en
 license: apache-2.0
 base_model: gokulsrinivasagan/tinybert_base_train_kd
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: tinybert_base_train_kd_qqp
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QQP
+      type: glue
+      args: qqp
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8704922087558743
+    - name: F1
+      type: f1
+      value: 0.8217592592592593
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -17,12 +35,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tinybert_base_train_kd_qqp
 
-This model is a fine-tuned version of [gokulsrinivasagan/tinybert_base_train_kd](https://huggingface.co/gokulsrinivasagan/tinybert_base_train_kd) on an unknown dataset.
+This model is a fine-tuned version of [gokulsrinivasagan/tinybert_base_train_kd](https://huggingface.co/gokulsrinivasagan/tinybert_base_train_kd) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4351
-- Accuracy: 0.8796
-- F1: 0.8394
-- Combined Score: 0.8595
+- Loss: 0.3005
+- Accuracy: 0.8705
+- F1: 0.8218
+- Combined Score: 0.8461
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.8704922087558743,
+    "eval_combined_score": 0.8461257340075667,
+    "eval_f1": 0.8217592592592593,
+    "eval_loss": 0.30050283670425415,
+    "eval_runtime": 19.9759,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 2023.941,
+    "eval_steps_per_second": 7.91,
+    "total_flos": 2.168897990912594e+17,
+    "train_loss": 0.21321062558068765,
+    "train_runtime": 3308.4254,
+    "train_samples": 363846,
+    "train_samples_per_second": 5498.779,
+    "train_steps_per_second": 21.491
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.8704922087558743,
+    "eval_combined_score": 0.8461257340075667,
+    "eval_f1": 0.8217592592592593,
+    "eval_loss": 0.30050283670425415,
+    "eval_runtime": 19.9759,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 2023.941,
+    "eval_steps_per_second": 7.91
+}

logs/events.out.tfevents.1744838373.ki-g0008.514628.25

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c531726b7b37352945aaf9f1d2a4c9830fca602be9a783a891783d10dcd55bbc
+size 515

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 2.168897990912594e+17,
+    "train_loss": 0.21321062558068765,
+    "train_runtime": 3308.4254,
+    "train_samples": 363846,
+    "train_samples_per_second": 5498.779,
+    "train_steps_per_second": 21.491
+}

trainer_state.json

@@ -0,0 +1,214 @@
+{
+  "best_global_step": 5688,
+  "best_metric": 0.30050283670425415,
+  "best_model_checkpoint": "tinybert_base_train_kd_qqp/checkpoint-5688",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 12798,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.605546474456787,
+      "learning_rate": 4.900070323488045e-05,
+      "loss": 0.4061,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8430373485035865,
+      "eval_combined_score": 0.8128542906901495,
+      "eval_f1": 0.7826712328767124,
+      "eval_loss": 0.34433236718177795,
+      "eval_runtime": 19.9883,
+      "eval_samples_per_second": 2022.682,
+      "eval_steps_per_second": 7.905,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.7136270999908447,
+      "learning_rate": 4.800070323488045e-05,
+      "loss": 0.3179,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8611427158050953,
+      "eval_combined_score": 0.8380532300185661,
+      "eval_f1": 0.8149637442320369,
+      "eval_loss": 0.3137979209423065,
+      "eval_runtime": 20.1369,
+      "eval_samples_per_second": 2007.752,
+      "eval_steps_per_second": 7.846,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.9198460578918457,
+      "learning_rate": 4.700070323488045e-05,
+      "loss": 0.269,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8663368785555281,
+      "eval_combined_score": 0.8460577293673894,
+      "eval_f1": 0.8257785801792508,
+      "eval_loss": 0.3077119290828705,
+      "eval_runtime": 20.3208,
+      "eval_samples_per_second": 1989.585,
+      "eval_steps_per_second": 7.775,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.6862900257110596,
+      "learning_rate": 4.600070323488046e-05,
+      "loss": 0.2275,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8704922087558743,
+      "eval_combined_score": 0.8461257340075667,
+      "eval_f1": 0.8217592592592593,
+      "eval_loss": 0.30050283670425415,
+      "eval_runtime": 19.99,
+      "eval_samples_per_second": 2022.51,
+      "eval_steps_per_second": 7.904,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.6690852642059326,
+      "learning_rate": 4.500070323488045e-05,
+      "loss": 0.1923,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8749690823645807,
+      "eval_combined_score": 0.8502059749334558,
+      "eval_f1": 0.8254428675023309,
+      "eval_loss": 0.3269473910331726,
+      "eval_runtime": 20.1113,
+      "eval_samples_per_second": 2010.309,
+      "eval_steps_per_second": 7.856,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.443815231323242,
+      "learning_rate": 4.400070323488045e-05,
+      "loss": 0.1614,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8750432846895869,
+      "eval_combined_score": 0.8520798888608265,
+      "eval_f1": 0.829116493032066,
+      "eval_loss": 0.3224050998687744,
+      "eval_runtime": 20.1576,
+      "eval_samples_per_second": 2005.698,
+      "eval_steps_per_second": 7.838,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.2654778957366943,
+      "learning_rate": 4.300070323488045e-05,
+      "loss": 0.1349,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8761068513480089,
+      "eval_combined_score": 0.857411380402407,
+      "eval_f1": 0.8387159094568052,
+      "eval_loss": 0.3559403121471405,
+      "eval_runtime": 20.0974,
+      "eval_samples_per_second": 2011.705,
+      "eval_steps_per_second": 7.862,
+      "step": 9954
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.175635814666748,
+      "learning_rate": 4.200070323488045e-05,
+      "loss": 0.1132,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8775414296314618,
+      "eval_combined_score": 0.8549946141016627,
+      "eval_f1": 0.8324477985718637,
+      "eval_loss": 0.4348158538341522,
+      "eval_runtime": 19.998,
+      "eval_samples_per_second": 2021.701,
+      "eval_steps_per_second": 7.901,
+      "step": 11376
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 3.47190523147583,
+      "learning_rate": 4.100070323488045e-05,
+      "loss": 0.0966,
+      "step": 12798
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.8796438288399703,
+      "eval_combined_score": 0.8595354835857718,
+      "eval_f1": 0.8394271383315733,
+      "eval_loss": 0.435083270072937,
+      "eval_runtime": 20.1566,
+      "eval_samples_per_second": 2005.79,
+      "eval_steps_per_second": 7.839,
+      "step": 12798
+    },
+    {
+      "epoch": 9.0,
+      "step": 12798,
+      "total_flos": 2.168897990912594e+17,
+      "train_loss": 0.21321062558068765,
+      "train_runtime": 3308.4254,
+      "train_samples_per_second": 5498.779,
+      "train_steps_per_second": 21.491
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 71100,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 5
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.168897990912594e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}