End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_20_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_base_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: 0.6459519934906428
 ---
 
 <!-- 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_20_v1_mnli
 
-This model is a fine-tuned version of [Hartunka/bert_base_km_20_v1](https://huggingface.co/Hartunka/bert_base_km_20_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_km_20_v1](https://huggingface.co/Hartunka/bert_base_km_20_v1) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1190
-- Accuracy: 0.6398
+- Loss: 0.8045
+- Accuracy: 0.6460
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 8.0,
+    "epoch_mm": 8.0,
+    "eval_accuracy": 0.6453387671930718,
+    "eval_accuracy_mm": 0.6459519934906428,
+    "eval_loss": 0.8076121211051941,
+    "eval_loss_mm": 0.8044986724853516,
+    "eval_runtime": 6.1223,
+    "eval_runtime_mm": 6.1909,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 1603.146,
+    "eval_samples_per_second_mm": 1588.144,
+    "eval_steps_per_second": 6.37,
+    "eval_steps_per_second_mm": 6.3,
+    "total_flos": 4.133006614744842e+17,
+    "train_loss": 0.6673230117738169,
+    "train_runtime": 5085.7727,
+    "train_samples": 392702,
+    "train_samples_per_second": 3860.79,
+    "train_steps_per_second": 15.081
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "epoch_mm": 8.0,
+    "eval_accuracy": 0.6453387671930718,
+    "eval_accuracy_mm": 0.6459519934906428,
+    "eval_loss": 0.8076121211051941,
+    "eval_loss_mm": 0.8044986724853516,
+    "eval_runtime": 6.1223,
+    "eval_runtime_mm": 6.1909,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 1603.146,
+    "eval_samples_per_second_mm": 1588.144,
+    "eval_steps_per_second": 6.37,
+    "eval_steps_per_second_mm": 6.3
+}

logs/events.out.tfevents.1745055933.s_005_m.2788007.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85f77e8bba491a2043de1f68cf7a3c3d0c6b15549c09efc805fc9cd7a0e6340e
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 4.133006614744842e+17,
+    "train_loss": 0.6673230117738169,
+    "train_runtime": 5085.7727,
+    "train_samples": 392702,
+    "train_samples_per_second": 3860.79,
+    "train_steps_per_second": 15.081
+}

trainer_state.json

@@ -0,0 +1,180 @@
+{
+  "best_global_step": 4602,
+  "best_metric": 0.8076121211051941,
+  "best_model_checkpoint": "bert_base_km_20_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": 1.2500191926956177,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9865,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5676006113092206,
+      "eval_loss": 0.9101185202598572,
+      "eval_runtime": 6.2505,
+      "eval_samples_per_second": 1570.283,
+      "eval_steps_per_second": 6.24,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.4807652235031128,
+      "learning_rate": 4.8e-05,
+      "loss": 0.8711,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6211920529801325,
+      "eval_loss": 0.8458642363548279,
+      "eval_runtime": 6.1413,
+      "eval_samples_per_second": 1598.207,
+      "eval_steps_per_second": 6.35,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.847516655921936,
+      "learning_rate": 4.7e-05,
+      "loss": 0.7857,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6453387671930718,
+      "eval_loss": 0.8076121211051941,
+      "eval_runtime": 6.1529,
+      "eval_samples_per_second": 1595.189,
+      "eval_steps_per_second": 6.338,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.9729530811309814,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.7044,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6398369842078452,
+      "eval_loss": 0.8136552572250366,
+      "eval_runtime": 6.153,
+      "eval_samples_per_second": 1595.145,
+      "eval_steps_per_second": 6.338,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.507617712020874,
+      "learning_rate": 4.5e-05,
+      "loss": 0.6239,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6522669383596535,
+      "eval_loss": 0.8299856185913086,
+      "eval_runtime": 6.1783,
+      "eval_samples_per_second": 1588.614,
+      "eval_steps_per_second": 6.312,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.8399038314819336,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.5378,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6479877738155884,
+      "eval_loss": 0.9015277624130249,
+      "eval_runtime": 6.1622,
+      "eval_samples_per_second": 1592.765,
+      "eval_steps_per_second": 6.329,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.285945177078247,
+      "learning_rate": 4.3e-05,
+      "loss": 0.4537,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6478858889454916,
+      "eval_loss": 0.9738817811012268,
+      "eval_runtime": 6.1389,
+      "eval_samples_per_second": 1598.83,
+      "eval_steps_per_second": 6.353,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 3.6435606479644775,
+      "learning_rate": 4.2e-05,
+      "loss": 0.3755,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6398369842078452,
+      "eval_loss": 1.1189723014831543,
+      "eval_runtime": 6.1437,
+      "eval_samples_per_second": 1597.584,
+      "eval_steps_per_second": 6.348,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "step": 12272,
+      "total_flos": 4.133006614744842e+17,
+      "train_loss": 0.6673230117738169,
+      "train_runtime": 5085.7727,
+      "train_samples_per_second": 3860.79,
+      "train_steps_per_second": 15.081
+    }
+  ],
+  "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
+}