End of training

README.md

@@ -1,15 +1,33 @@
 ---
 library_name: transformers
+language:
+- en
 license: apache-2.0
 base_model: gokulsrinivasagan/tinybert_train_book_ent_15p_mid
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: tinybert_train_book_ent_15p_mid_mrpc
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MRPC
+      type: glue
+      args: mrpc
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6838235294117647
+    - name: F1
+      type: f1
+      value: 0.8122270742358079
 ---
 
 <!-- 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_train_book_ent_15p_mid_mrpc
 
-This model is a fine-tuned version of [gokulsrinivasagan/tinybert_train_book_ent_15p_mid](https://huggingface.co/gokulsrinivasagan/tinybert_train_book_ent_15p_mid) on an unknown dataset.
+This model is a fine-tuned version of [gokulsrinivasagan/tinybert_train_book_ent_15p_mid](https://huggingface.co/gokulsrinivasagan/tinybert_train_book_ent_15p_mid) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0058
-- Accuracy: 0.7108
-- F1: 0.8115
-- Combined Score: 0.7611
+- Loss: 0.6241
+- Accuracy: 0.6838
+- F1: 0.8122
+- Combined Score: 0.7480
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6838235294117647,
+    "eval_combined_score": 0.7480253018237863,
+    "eval_f1": 0.8122270742358079,
+    "eval_loss": 0.6241479516029358,
+    "eval_runtime": 0.206,
+    "eval_samples": 408,
+    "eval_samples_per_second": 1980.16,
+    "eval_steps_per_second": 9.707,
+    "total_flos": 769504676118528.0,
+    "train_loss": 0.4986685355504354,
+    "train_runtime": 27.9584,
+    "train_samples": 3668,
+    "train_samples_per_second": 6559.737,
+    "train_steps_per_second": 26.826
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6838235294117647,
+    "eval_combined_score": 0.7480253018237863,
+    "eval_f1": 0.8122270742358079,
+    "eval_loss": 0.6241479516029358,
+    "eval_runtime": 0.206,
+    "eval_samples": 408,
+    "eval_samples_per_second": 1980.16,
+    "eval_steps_per_second": 9.707
+}

logs/events.out.tfevents.1744842862.ki-g0008.514628.39

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e3b9e3b306ab1efc00045d10865ba4ba1d92f22b63b3afb256e0718e77dc908
+size 507

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 769504676118528.0,
+    "train_loss": 0.4986685355504354,
+    "train_runtime": 27.9584,
+    "train_samples": 3668,
+    "train_samples_per_second": 6559.737,
+    "train_steps_per_second": 26.826
+}

trainer_state.json

@@ -0,0 +1,196 @@
+{
+  "best_global_step": 45,
+  "best_metric": 0.6241479516029358,
+  "best_model_checkpoint": "tinybert_train_book_ent_15p_mid_mrpc/checkpoint-45",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.45011627674102783,
+      "learning_rate": 4.906666666666667e-05,
+      "loss": 0.6351,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6255457401275635,
+      "eval_runtime": 0.2193,
+      "eval_samples_per_second": 1860.613,
+      "eval_steps_per_second": 9.121,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.5221871137619019,
+      "learning_rate": 4.806666666666667e-05,
+      "loss": 0.6347,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6243202090263367,
+      "eval_runtime": 0.1987,
+      "eval_samples_per_second": 2053.352,
+      "eval_steps_per_second": 10.065,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.8151041269302368,
+      "learning_rate": 4.706666666666667e-05,
+      "loss": 0.6143,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6241479516029358,
+      "eval_runtime": 0.2182,
+      "eval_samples_per_second": 1869.908,
+      "eval_steps_per_second": 9.166,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.2786083221435547,
+      "learning_rate": 4.606666666666667e-05,
+      "loss": 0.5916,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.571078431372549,
+      "eval_combined_score": 0.59897205150717,
+      "eval_f1": 0.6268656716417911,
+      "eval_loss": 0.6595872044563293,
+      "eval_runtime": 0.1901,
+      "eval_samples_per_second": 2145.868,
+      "eval_steps_per_second": 10.519,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.2732080221176147,
+      "learning_rate": 4.5066666666666667e-05,
+      "loss": 0.5301,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7517156862745098,
+      "eval_f1": 0.8,
+      "eval_loss": 0.6268830895423889,
+      "eval_runtime": 0.2139,
+      "eval_samples_per_second": 1907.335,
+      "eval_steps_per_second": 9.35,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.419187307357788,
+      "learning_rate": 4.406666666666667e-05,
+      "loss": 0.4009,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7339222378371144,
+      "eval_f1": 0.7815699658703071,
+      "eval_loss": 0.8087877035140991,
+      "eval_runtime": 0.1937,
+      "eval_samples_per_second": 2106.653,
+      "eval_steps_per_second": 10.327,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 6.955132484436035,
+      "learning_rate": 4.3066666666666665e-05,
+      "loss": 0.3157,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.678921568627451,
+      "eval_combined_score": 0.727876457226843,
+      "eval_f1": 0.7768313458262351,
+      "eval_loss": 0.9177660346031189,
+      "eval_runtime": 0.2085,
+      "eval_samples_per_second": 1957.064,
+      "eval_steps_per_second": 9.593,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 5.208475112915039,
+      "learning_rate": 4.206666666666667e-05,
+      "loss": 0.2669,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7107843137254902,
+      "eval_combined_score": 0.7611429555847898,
+      "eval_f1": 0.8115015974440895,
+      "eval_loss": 1.0057904720306396,
+      "eval_runtime": 0.2033,
+      "eval_samples_per_second": 2007.296,
+      "eval_steps_per_second": 9.84,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "step": 120,
+      "total_flos": 769504676118528.0,
+      "train_loss": 0.4986685355504354,
+      "train_runtime": 27.9584,
+      "train_samples_per_second": 6559.737,
+      "train_steps_per_second": 26.826
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 750,
+  "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": 769504676118528.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}