End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_5_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_base_km_5_v2_mnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.684499593165175
 ---
 
 <!-- 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. -->
 
 # bert_base_km_5_v2_mnli
 
-This model is a fine-tuned version of [Hartunka/bert_base_km_5_v2](https://huggingface.co/Hartunka/bert_base_km_5_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_km_5_v2](https://huggingface.co/Hartunka/bert_base_km_5_v2) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0465
-- Accuracy: 0.6881
+- Loss: 0.7223
+- Accuracy: 0.6845
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 8.0,
+    "epoch_mm": 8.0,
+    "eval_accuracy": 0.6837493632195619,
+    "eval_accuracy_mm": 0.684499593165175,
+    "eval_loss": 0.7321322560310364,
+    "eval_loss_mm": 0.722257673740387,
+    "eval_runtime": 6.1094,
+    "eval_runtime_mm": 6.1393,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 1606.552,
+    "eval_samples_per_second_mm": 1601.488,
+    "eval_steps_per_second": 6.384,
+    "eval_steps_per_second_mm": 6.353,
+    "total_flos": 4.133006614744842e+17,
+    "train_loss": 0.5957969854521472,
+    "train_runtime": 5091.5013,
+    "train_samples": 392702,
+    "train_samples_per_second": 3856.446,
+    "train_steps_per_second": 15.064
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "epoch_mm": 8.0,
+    "eval_accuracy": 0.6837493632195619,
+    "eval_accuracy_mm": 0.684499593165175,
+    "eval_loss": 0.7321322560310364,
+    "eval_loss_mm": 0.722257673740387,
+    "eval_runtime": 6.1094,
+    "eval_runtime_mm": 6.1393,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 1606.552,
+    "eval_samples_per_second_mm": 1601.488,
+    "eval_steps_per_second": 6.384,
+    "eval_steps_per_second_mm": 6.353
+}

logs/events.out.tfevents.1745280156.s_005_m.2836894.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3edc18debbc2951b34b88be9526aceea324b0eada025cae4b2a5834df7caa21
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 4.133006614744842e+17,
+    "train_loss": 0.5957969854521472,
+    "train_runtime": 5091.5013,
+    "train_samples": 392702,
+    "train_samples_per_second": 3856.446,
+    "train_steps_per_second": 15.064
+}

trainer_state.json

@@ -0,0 +1,180 @@
+{
+  "best_global_step": 4602,
+  "best_metric": 0.7321322560310364,
+  "best_model_checkpoint": "bert_base_km_5_v2_mnli/checkpoint-4602",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 12272,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.838555932044983,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9698,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.602445236882323,
+      "eval_loss": 0.88103848695755,
+      "eval_runtime": 6.1863,
+      "eval_samples_per_second": 1586.562,
+      "eval_steps_per_second": 6.304,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.7681314945220947,
+      "learning_rate": 4.8e-05,
+      "loss": 0.8146,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6581762608252675,
+      "eval_loss": 0.7751404643058777,
+      "eval_runtime": 6.1214,
+      "eval_samples_per_second": 1603.403,
+      "eval_steps_per_second": 6.371,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.8645142316818237,
+      "learning_rate": 4.7e-05,
+      "loss": 0.7087,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6837493632195619,
+      "eval_loss": 0.7321322560310364,
+      "eval_runtime": 6.2815,
+      "eval_samples_per_second": 1562.515,
+      "eval_steps_per_second": 6.209,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.204162836074829,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.6219,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6838512480896587,
+      "eval_loss": 0.7466456890106201,
+      "eval_runtime": 6.1611,
+      "eval_samples_per_second": 1593.064,
+      "eval_steps_per_second": 6.33,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.40364408493042,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5372,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6897605705552725,
+      "eval_loss": 0.7687764763832092,
+      "eval_runtime": 6.2669,
+      "eval_samples_per_second": 1566.158,
+      "eval_steps_per_second": 6.223,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.0168914794921875,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4493,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6916963830871116,
+      "eval_loss": 0.8502929210662842,
+      "eval_runtime": 6.196,
+      "eval_samples_per_second": 1584.08,
+      "eval_steps_per_second": 6.294,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.2449002265930176,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3668,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.689556800815079,
+      "eval_loss": 0.9104171395301819,
+      "eval_runtime": 6.2266,
+      "eval_samples_per_second": 1576.309,
+      "eval_steps_per_second": 6.263,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.044001579284668,
+      "learning_rate": 4.2e-05,
+      "loss": 0.298,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6881304126337239,
+      "eval_loss": 1.0464503765106201,
+      "eval_runtime": 6.1342,
+      "eval_samples_per_second": 1600.034,
+      "eval_steps_per_second": 6.358,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "step": 12272,
+      "total_flos": 4.133006614744842e+17,
+      "train_loss": 0.5957969854521472,
+      "train_runtime": 5091.5013,
+      "train_samples_per_second": 3856.446,
+      "train_steps_per_second": 15.064
+    }
+  ],
+  "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": 4.133006614744842e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}