End of training

README.md

@@ -1,14 +1,29 @@
 ---
 library_name: transformers
+language:
+- en
 license: apache-2.0
 base_model: google/bert_uncased_L-4_H-256_A-4
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_uncased_L-4_H-256_A-4_mnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.7651545972335232
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_uncased_L-4_H-256_A-4_mnli
 
-This model is a fine-tuned version of [google/bert_uncased_L-4_H-256_A-4](https://huggingface.co/google/bert_uncased_L-4_H-256_A-4) on an unknown dataset.
+This model is a fine-tuned version of [google/bert_uncased_L-4_H-256_A-4](https://huggingface.co/google/bert_uncased_L-4_H-256_A-4) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6584
-- Accuracy: 0.7608
+- Loss: 0.5852
+- Accuracy: 0.7652
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.7552725420275089,
+    "eval_accuracy_mm": 0.7651545972335232,
+    "eval_loss": 0.6099433302879333,
+    "eval_loss_mm": 0.5852423310279846,
+    "eval_runtime": 3.4113,
+    "eval_runtime_mm": 3.4446,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 2877.217,
+    "eval_samples_per_second_mm": 2854.319,
+    "eval_steps_per_second": 11.433,
+    "eval_steps_per_second_mm": 11.322,
+    "total_flos": 1.751365065917952e+16,
+    "train_loss": 0.5555856487258144,
+    "train_runtime": 1686.0242,
+    "train_samples": 392702,
+    "train_samples_per_second": 11645.8,
+    "train_steps_per_second": 45.492
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.7552725420275089,
+    "eval_accuracy_mm": 0.7651545972335232,
+    "eval_loss": 0.6099433302879333,
+    "eval_loss_mm": 0.5852423310279846,
+    "eval_runtime": 3.4113,
+    "eval_runtime_mm": 3.4446,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 2877.217,
+    "eval_samples_per_second_mm": 2854.319,
+    "eval_steps_per_second": 11.433,
+    "eval_steps_per_second_mm": 11.322
+}

logs/events.out.tfevents.1733334846.ki-g0008.1761130.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:406e4e4ad18e425c9abe356ced792229b427b2115468124db1665432e613d153
+size 734

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 1.751365065917952e+16,
+    "train_loss": 0.5555856487258144,
+    "train_runtime": 1686.0242,
+    "train_samples": 392702,
+    "train_samples_per_second": 11645.8,
+    "train_steps_per_second": 45.492
+}

trainer_state.json

@@ -0,0 +1,195 @@
+{
+  "best_metric": 0.6099433302879333,
+  "best_model_checkpoint": "bert_uncased_L-4_H-256_A-4_mnli/checkpoint-6136",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 13806,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.477856159210205,
+      "learning_rate": 4.9e-05,
+      "loss": 0.7878,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7006622516556291,
+      "eval_loss": 0.7087482213973999,
+      "eval_runtime": 3.4685,
+      "eval_samples_per_second": 2829.716,
+      "eval_steps_per_second": 11.244,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.7286040782928467,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6683,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7295975547631177,
+      "eval_loss": 0.6437011361122131,
+      "eval_runtime": 3.5697,
+      "eval_samples_per_second": 2749.523,
+      "eval_steps_per_second": 10.925,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.3989334106445312,
+      "learning_rate": 4.7e-05,
+      "loss": 0.6112,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7465104431991849,
+      "eval_loss": 0.6203623414039612,
+      "eval_runtime": 3.4599,
+      "eval_samples_per_second": 2836.766,
+      "eval_steps_per_second": 11.272,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.6142916679382324,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5683,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7552725420275089,
+      "eval_loss": 0.6099433302879333,
+      "eval_runtime": 3.4644,
+      "eval_samples_per_second": 2833.094,
+      "eval_steps_per_second": 11.257,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.1504721641540527,
+      "learning_rate": 4.5e-05,
+      "loss": 0.532,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.757208354559348,
+      "eval_loss": 0.6147148013114929,
+      "eval_runtime": 3.3611,
+      "eval_samples_per_second": 2920.179,
+      "eval_steps_per_second": 11.603,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.1118323802948,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4997,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7551706571574122,
+      "eval_loss": 0.6381344199180603,
+      "eval_runtime": 3.4815,
+      "eval_samples_per_second": 2819.177,
+      "eval_steps_per_second": 11.202,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.8937153816223145,
+      "learning_rate": 4.3e-05,
+      "loss": 0.4707,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7588385124808966,
+      "eval_loss": 0.6195704340934753,
+      "eval_runtime": 3.458,
+      "eval_samples_per_second": 2838.345,
+      "eval_steps_per_second": 11.278,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 3.1744437217712402,
+      "learning_rate": 4.2e-05,
+      "loss": 0.4436,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7589403973509934,
+      "eval_loss": 0.6404411196708679,
+      "eval_runtime": 3.4687,
+      "eval_samples_per_second": 2829.597,
+      "eval_steps_per_second": 11.243,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.90157151222229,
+      "learning_rate": 4.1e-05,
+      "loss": 0.4187,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7607743250127356,
+      "eval_loss": 0.6584197282791138,
+      "eval_runtime": 3.4328,
+      "eval_samples_per_second": 2859.191,
+      "eval_steps_per_second": 11.361,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "step": 13806,
+      "total_flos": 1.751365065917952e+16,
+      "train_loss": 0.5555856487258144,
+      "train_runtime": 1686.0242,
+      "train_samples_per_second": 11645.8,
+      "train_steps_per_second": 45.492
+    }
+  ],
+  "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": 1.751365065917952e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}