End of training

README.md

@@ -1,13 +1,31 @@
 ---
+language:
+- en
 base_model: Hartunka/tiny_bert_km_100_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: tiny_bert_km_100_v1_mrpc
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MRPC
+      type: glue
+      args: mrpc
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.7230392156862745
+    - name: F1
+      type: f1
+      value: 0.8197767145135566
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,12 +33,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tiny_bert_km_100_v1_mrpc
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_100_v1](https://huggingface.co/Hartunka/tiny_bert_km_100_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_100_v1](https://huggingface.co/Hartunka/tiny_bert_km_100_v1) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9716
-- Accuracy: 0.6593
-- F1: 0.7440
-- Combined Score: 0.7017
+- Loss: 0.5885
+- Accuracy: 0.7230
+- F1: 0.8198
+- Combined Score: 0.7714
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.7230392156862745,
+    "eval_combined_score": 0.7714079650999155,
+    "eval_f1": 0.8197767145135566,
+    "eval_loss": 0.5884530544281006,
+    "eval_runtime": 0.1883,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2166.745,
+    "eval_steps_per_second": 10.621,
+    "total_flos": 865692760633344.0,
+    "train_loss": 0.4522992098772967,
+    "train_runtime": 30.6833,
+    "train_samples": 3668,
+    "train_samples_per_second": 5977.188,
+    "train_steps_per_second": 24.443
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.7230392156862745,
+    "eval_combined_score": 0.7714079650999155,
+    "eval_f1": 0.8197767145135566,
+    "eval_loss": 0.5884530544281006,
+    "eval_runtime": 0.1883,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2166.745,
+    "eval_steps_per_second": 10.621
+}

logs/events.out.tfevents.1744818268.s_004_m

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:086095e74c9dc040f9763b1b96e1bab769f2fe39cb7dc0f62a37ecafd9d89546
+size 40

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 865692760633344.0,
+    "train_loss": 0.4522992098772967,
+    "train_runtime": 30.6833,
+    "train_samples": 3668,
+    "train_samples_per_second": 5977.188,
+    "train_steps_per_second": 24.443
+}

trainer_state.json

@@ -0,0 +1,192 @@
+{
+  "best_metric": 0.5884530544281006,
+  "best_model_checkpoint": "tiny_bert_km_100_v1_mrpc/checkpoint-60",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 135,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.8016844987869263,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6262,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.696078431372549,
+      "eval_combined_score": 0.7511642156862746,
+      "eval_f1": 0.80625,
+      "eval_loss": 0.6085503697395325,
+      "eval_runtime": 0.2153,
+      "eval_samples_per_second": 1894.685,
+      "eval_steps_per_second": 9.288,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.281128168106079,
+      "learning_rate": 4.8e-05,
+      "loss": 0.599,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7612822034643454,
+      "eval_f1": 0.8191330343796711,
+      "eval_loss": 0.6000078320503235,
+      "eval_runtime": 0.1728,
+      "eval_samples_per_second": 2360.938,
+      "eval_steps_per_second": 11.573,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.0308752059936523,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5725,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_combined_score": 0.7617557661362417,
+      "eval_f1": 0.817629179331307,
+      "eval_loss": 0.5969370007514954,
+      "eval_runtime": 0.1629,
+      "eval_samples_per_second": 2504.949,
+      "eval_steps_per_second": 12.279,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.1744054555892944,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5494,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7230392156862745,
+      "eval_combined_score": 0.7714079650999155,
+      "eval_f1": 0.8197767145135566,
+      "eval_loss": 0.5884530544281006,
+      "eval_runtime": 0.1682,
+      "eval_samples_per_second": 2426.402,
+      "eval_steps_per_second": 11.894,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.2589306831359863,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5073,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7335482138060704,
+      "eval_f1": 0.7808219178082192,
+      "eval_loss": 0.6107202172279358,
+      "eval_runtime": 0.1757,
+      "eval_samples_per_second": 2321.618,
+      "eval_steps_per_second": 11.38,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.4866365194320679,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4408,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.7474496555378909,
+      "eval_f1": 0.793918918918919,
+      "eval_loss": 0.636547327041626,
+      "eval_runtime": 0.1723,
+      "eval_samples_per_second": 2367.506,
+      "eval_steps_per_second": 11.605,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.2932753562927246,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3565,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7107843137254902,
+      "eval_combined_score": 0.7560655575361458,
+      "eval_f1": 0.8013468013468014,
+      "eval_loss": 0.7094822525978088,
+      "eval_runtime": 0.1637,
+      "eval_samples_per_second": 2492.333,
+      "eval_steps_per_second": 12.217,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.5407330989837646,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2434,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7486492467515116,
+      "eval_f1": 0.7938671209540034,
+      "eval_loss": 0.8136605024337769,
+      "eval_runtime": 0.1568,
+      "eval_samples_per_second": 2602.642,
+      "eval_steps_per_second": 12.758,
+      "step": 120
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 3.441915512084961,
+      "learning_rate": 4.1e-05,
+      "loss": 0.1755,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6593137254901961,
+      "eval_combined_score": 0.7016642292276026,
+      "eval_f1": 0.7440147329650092,
+      "eval_loss": 0.9716194272041321,
+      "eval_runtime": 0.1674,
+      "eval_samples_per_second": 2437.535,
+      "eval_steps_per_second": 11.949,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "step": 135,
+      "total_flos": 865692760633344.0,
+      "train_loss": 0.4522992098772967,
+      "train_runtime": 30.6833,
+      "train_samples_per_second": 5977.188,
+      "train_steps_per_second": 24.443
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 750,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "total_flos": 865692760633344.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}