End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_50_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_base_km_50_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: 0.6587672904800651
 ---
 
 <!-- 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_50_v1_mnli
 
-This model is a fine-tuned version of [Hartunka/bert_base_km_50_v1](https://huggingface.co/Hartunka/bert_base_km_50_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_km_50_v1](https://huggingface.co/Hartunka/bert_base_km_50_v1) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0233
-- Accuracy: 0.6580
+- Loss: 0.7719
+- Accuracy: 0.6588
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 8.0,
+    "epoch_mm": 8.0,
+    "eval_accuracy": 0.6573611818644931,
+    "eval_accuracy_mm": 0.6587672904800651,
+    "eval_loss": 0.7822551727294922,
+    "eval_loss_mm": 0.7719444632530212,
+    "eval_runtime": 6.166,
+    "eval_runtime_mm": 6.3434,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 1591.803,
+    "eval_samples_per_second_mm": 1549.949,
+    "eval_steps_per_second": 6.325,
+    "eval_steps_per_second_mm": 6.148,
+    "total_flos": 4.133006614744842e+17,
+    "train_loss": 0.6634338806629803,
+    "train_runtime": 5096.162,
+    "train_samples": 392702,
+    "train_samples_per_second": 3852.919,
+    "train_steps_per_second": 15.051
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "epoch_mm": 8.0,
+    "eval_accuracy": 0.6573611818644931,
+    "eval_accuracy_mm": 0.6587672904800651,
+    "eval_loss": 0.7822551727294922,
+    "eval_loss_mm": 0.7719444632530212,
+    "eval_runtime": 6.166,
+    "eval_runtime_mm": 6.3434,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 1591.803,
+    "eval_samples_per_second_mm": 1549.949,
+    "eval_steps_per_second": 6.325,
+    "eval_steps_per_second_mm": 6.148
+}

logs/events.out.tfevents.1745108358.s_005_m.2796520.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4b009baf3df7c7b7b5023be83781477c84f19799a257e1b1b5e472c1efd402d
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 4.133006614744842e+17,
+    "train_loss": 0.6634338806629803,
+    "train_runtime": 5096.162,
+    "train_samples": 392702,
+    "train_samples_per_second": 3852.919,
+    "train_steps_per_second": 15.051
+}

trainer_state.json

@@ -0,0 +1,180 @@
+{
+  "best_global_step": 4602,
+  "best_metric": 0.7822551727294922,
+  "best_model_checkpoint": "bert_base_km_50_v1_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": 1.0520974397659302,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9861,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5751400916963831,
+      "eval_loss": 0.906815230846405,
+      "eval_runtime": 6.2844,
+      "eval_samples_per_second": 1561.793,
+      "eval_steps_per_second": 6.206,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.6486469507217407,
+      "learning_rate": 4.8e-05,
+      "loss": 0.8643,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6255731023942944,
+      "eval_loss": 0.8278823494911194,
+      "eval_runtime": 6.1983,
+      "eval_samples_per_second": 1583.511,
+      "eval_steps_per_second": 6.292,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.6458042860031128,
+      "learning_rate": 4.7e-05,
+      "loss": 0.7687,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6573611818644931,
+      "eval_loss": 0.7822551727294922,
+      "eval_runtime": 6.2005,
+      "eval_samples_per_second": 1582.936,
+      "eval_steps_per_second": 6.29,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.1764519214630127,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.6904,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6566479877738156,
+      "eval_loss": 0.7888372540473938,
+      "eval_runtime": 6.1797,
+      "eval_samples_per_second": 1588.259,
+      "eval_steps_per_second": 6.311,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.208383798599243,
+      "learning_rate": 4.5e-05,
+      "loss": 0.6157,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6703005603667855,
+      "eval_loss": 0.8072159886360168,
+      "eval_runtime": 6.1666,
+      "eval_samples_per_second": 1591.641,
+      "eval_steps_per_second": 6.324,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.21551775932312,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.5387,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6684666327050433,
+      "eval_loss": 0.8871552348136902,
+      "eval_runtime": 6.1896,
+      "eval_samples_per_second": 1585.734,
+      "eval_steps_per_second": 6.301,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.908752918243408,
+      "learning_rate": 4.3e-05,
+      "loss": 0.4595,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6663270504330107,
+      "eval_loss": 0.9230899810791016,
+      "eval_runtime": 6.196,
+      "eval_samples_per_second": 1584.074,
+      "eval_steps_per_second": 6.294,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 3.3069233894348145,
+      "learning_rate": 4.2e-05,
+      "loss": 0.3841,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6579724910850738,
+      "eval_loss": 1.0233361721038818,
+      "eval_runtime": 6.1561,
+      "eval_samples_per_second": 1594.352,
+      "eval_steps_per_second": 6.335,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "step": 12272,
+      "total_flos": 4.133006614744842e+17,
+      "train_loss": 0.6634338806629803,
+      "train_runtime": 5096.162,
+      "train_samples_per_second": 3852.919,
+      "train_steps_per_second": 15.051
+    }
+  ],
+  "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
+}