End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_20_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_km_20_v1_mnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6186940602115542
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tiny_bert_km_20_v1_mnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_20_v1](https://huggingface.co/Hartunka/tiny_bert_km_20_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_20_v1](https://huggingface.co/Hartunka/tiny_bert_km_20_v1) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0450
-- Accuracy: 0.6133
+- Loss: 0.8507
+- Accuracy: 0.6187
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6207845134997453,
+    "eval_accuracy_mm": 0.6186940602115542,
+    "eval_loss": 0.8501138091087341,
+    "eval_loss_mm": 0.8507214188575745,
+    "eval_runtime": 2.8543,
+    "eval_runtime_mm": 2.9638,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3438.612,
+    "eval_samples_per_second_mm": 3317.369,
+    "eval_steps_per_second": 13.663,
+    "eval_steps_per_second_mm": 13.159,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7503844856749613,
+    "train_runtime": 2101.2347,
+    "train_samples": 392702,
+    "train_samples_per_second": 9344.553,
+    "train_steps_per_second": 36.502
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6207845134997453,
+    "eval_accuracy_mm": 0.6186940602115542,
+    "eval_loss": 0.8501138091087341,
+    "eval_loss_mm": 0.8507214188575745,
+    "eval_runtime": 2.8543,
+    "eval_runtime_mm": 2.9638,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3438.612,
+    "eval_samples_per_second_mm": 3317.369,
+    "eval_steps_per_second": 13.663,
+    "eval_steps_per_second_mm": 13.159
+}

logs/events.out.tfevents.1744642830.s_005_m.2701049.53

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3987f7b2dc4939bc6a707c617db06ddc0613428563f2a90a0669b9b367a037c4
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7503844856749613,
+    "train_runtime": 2101.2347,
+    "train_samples": 392702,
+    "train_samples_per_second": 9344.553,
+    "train_steps_per_second": 36.502
+}

trainer_state.json

@@ -0,0 +1,212 @@
+{
+  "best_global_step": 7670,
+  "best_metric": 0.8501138091087341,
+  "best_model_checkpoint": "tiny_bert_km_20_v1_mnli/checkpoint-7670",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 15340,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.0432511568069458,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9983,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5526235354049923,
+      "eval_loss": 0.9320745468139648,
+      "eval_runtime": 2.9584,
+      "eval_samples_per_second": 3317.72,
+      "eval_steps_per_second": 13.183,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.5364327430725098,
+      "learning_rate": 4.8e-05,
+      "loss": 0.9041,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5864493122771268,
+      "eval_loss": 0.8863005638122559,
+      "eval_runtime": 2.8932,
+      "eval_samples_per_second": 3392.4,
+      "eval_steps_per_second": 13.48,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.3565590381622314,
+      "learning_rate": 4.7e-05,
+      "loss": 0.8513,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6091696383087112,
+      "eval_loss": 0.8583081960678101,
+      "eval_runtime": 2.8763,
+      "eval_samples_per_second": 3412.37,
+      "eval_steps_per_second": 13.559,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.692923665046692,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.8052,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6135506877228731,
+      "eval_loss": 0.854619026184082,
+      "eval_runtime": 2.8833,
+      "eval_samples_per_second": 3404.115,
+      "eval_steps_per_second": 13.526,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.7422435283660889,
+      "learning_rate": 4.5e-05,
+      "loss": 0.7608,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6207845134997453,
+      "eval_loss": 0.8501138091087341,
+      "eval_runtime": 2.8844,
+      "eval_samples_per_second": 3402.82,
+      "eval_steps_per_second": 13.521,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.037548303604126,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.7188,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6223127865511972,
+      "eval_loss": 0.8914830684661865,
+      "eval_runtime": 2.8854,
+      "eval_samples_per_second": 3401.561,
+      "eval_steps_per_second": 13.516,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.9302098751068115,
+      "learning_rate": 4.3e-05,
+      "loss": 0.6779,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6253693326541009,
+      "eval_loss": 0.8998808264732361,
+      "eval_runtime": 2.8446,
+      "eval_samples_per_second": 3450.424,
+      "eval_steps_per_second": 13.71,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.0882489681243896,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6365,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6173204279164544,
+      "eval_loss": 0.9568306803703308,
+      "eval_runtime": 2.8983,
+      "eval_samples_per_second": 3386.471,
+      "eval_steps_per_second": 13.456,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.653881072998047,
+      "learning_rate": 4.1e-05,
+      "loss": 0.5959,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6152827305145185,
+      "eval_loss": 0.9905173182487488,
+      "eval_runtime": 2.9003,
+      "eval_samples_per_second": 3384.151,
+      "eval_steps_per_second": 13.447,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.764169216156006,
+      "learning_rate": 4e-05,
+      "loss": 0.5551,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6133469179826796,
+      "eval_loss": 1.0450348854064941,
+      "eval_runtime": 2.9008,
+      "eval_samples_per_second": 3383.588,
+      "eval_steps_per_second": 13.445,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "step": 15340,
+      "total_flos": 1.029836100834816e+17,
+      "train_loss": 0.7503844856749613,
+      "train_runtime": 2101.2347,
+      "train_samples_per_second": 9344.553,
+      "train_steps_per_second": 36.502
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 76700,
+  "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": 1.029836100834816e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}