End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_10_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_base_km_10_v1_qqp
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QQP
+      type: glue
+      args: qqp
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8168933959930744
+    - name: F1
+      type: f1
+      value: 0.7483085710400164
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,12 +34,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_base_km_10_v1_qqp
 
-This model is a fine-tuned version of [Hartunka/bert_base_km_10_v1](https://huggingface.co/Hartunka/bert_base_km_10_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_km_10_v1](https://huggingface.co/Hartunka/bert_base_km_10_v1) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6077
-- Accuracy: 0.8357
-- F1: 0.7748
-- Combined Score: 0.8052
+- Loss: 0.3991
+- Accuracy: 0.8169
+- F1: 0.7483
+- Combined Score: 0.7826
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.8168933959930744,
+    "eval_combined_score": 0.7826009835165454,
+    "eval_f1": 0.7483085710400164,
+    "eval_loss": 0.39908739924430847,
+    "eval_runtime": 25.4192,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1590.533,
+    "eval_steps_per_second": 6.216,
+    "total_flos": 3.3506166766980096e+17,
+    "train_loss": 0.24196342696668346,
+    "train_runtime": 4302.4345,
+    "train_samples": 363846,
+    "train_samples_per_second": 4228.373,
+    "train_steps_per_second": 16.526
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.8168933959930744,
+    "eval_combined_score": 0.7826009835165454,
+    "eval_f1": 0.7483085710400164,
+    "eval_loss": 0.39908739924430847,
+    "eval_runtime": 25.4192,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1590.533,
+    "eval_steps_per_second": 6.216
+}

logs/events.out.tfevents.1744983397.s_005_m.2772371.25

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aebb60cf6ee6a1001d0c02e6095e633f34b79433721480d653790c9ecf4bdc62
+size 467

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "total_flos": 3.3506166766980096e+17,
+    "train_loss": 0.24196342696668346,
+    "train_runtime": 4302.4345,
+    "train_samples": 363846,
+    "train_samples_per_second": 4228.373,
+    "train_steps_per_second": 16.526
+}

trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_global_step": 2844,
+  "best_metric": 0.39908739924430847,
+  "best_model_checkpoint": "bert_base_km_10_v1_qqp/checkpoint-2844",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 9954,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.8532532453536987,
+      "learning_rate": 4.9e-05,
+      "loss": 0.4755,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7899826861241652,
+      "eval_combined_score": 0.7309140483760923,
+      "eval_f1": 0.6718454106280193,
+      "eval_loss": 0.44356149435043335,
+      "eval_runtime": 25.799,
+      "eval_samples_per_second": 1567.112,
+      "eval_steps_per_second": 6.124,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.911086082458496,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3635,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8168933959930744,
+      "eval_combined_score": 0.7826009835165454,
+      "eval_f1": 0.7483085710400164,
+      "eval_loss": 0.39908739924430847,
+      "eval_runtime": 25.503,
+      "eval_samples_per_second": 1585.302,
+      "eval_steps_per_second": 6.195,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.2880516052246094,
+      "learning_rate": 4.7e-05,
+      "loss": 0.2764,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8274301261439525,
+      "eval_combined_score": 0.7966945001925022,
+      "eval_f1": 0.7659588742410519,
+      "eval_loss": 0.4213642179965973,
+      "eval_runtime": 25.414,
+      "eval_samples_per_second": 1590.855,
+      "eval_steps_per_second": 6.217,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.7092745304107666,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2061,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.83447934701954,
+      "eval_combined_score": 0.7993729692685413,
+      "eval_f1": 0.7642665915175426,
+      "eval_loss": 0.47107040882110596,
+      "eval_runtime": 25.2725,
+      "eval_samples_per_second": 1599.766,
+      "eval_steps_per_second": 6.252,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.767702341079712,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1551,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8337620578778135,
+      "eval_combined_score": 0.7942305672432244,
+      "eval_f1": 0.7546990766086353,
+      "eval_loss": 0.5268918871879578,
+      "eval_runtime": 25.4108,
+      "eval_samples_per_second": 1591.056,
+      "eval_steps_per_second": 6.218,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 5.443331718444824,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1209,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8357160524363096,
+      "eval_combined_score": 0.8038283599635194,
+      "eval_f1": 0.7719406674907293,
+      "eval_loss": 0.5602436065673828,
+      "eval_runtime": 25.2783,
+      "eval_samples_per_second": 1599.395,
+      "eval_steps_per_second": 6.25,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.5738558769226074,
+      "learning_rate": 4.3e-05,
+      "loss": 0.0961,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8356665842196389,
+      "eval_combined_score": 0.8052383897237023,
+      "eval_f1": 0.7748101952277657,
+      "eval_loss": 0.6076688170433044,
+      "eval_runtime": 25.4562,
+      "eval_samples_per_second": 1588.218,
+      "eval_steps_per_second": 6.207,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "step": 9954,
+      "total_flos": 3.3506166766980096e+17,
+      "train_loss": 0.24196342696668346,
+      "train_runtime": 4302.4345,
+      "train_samples_per_second": 4228.373,
+      "train_steps_per_second": 16.526
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 71100,
+  "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": 3.3506166766980096e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}