turkish-toxic-comment-classifier

README.md

@@ -5,6 +5,7 @@ base_model: dbmdz/bert-base-turkish-cased
 tags:
 - generated_from_trainer
 metrics:
+- accuracy
 - f1
 model-index:
 - name: results
@@ -18,8 +19,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [dbmdz/bert-base-turkish-cased](https://huggingface.co/dbmdz/bert-base-turkish-cased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5889
-- F1: 0.7477
+- Loss: 0.0296
+- Accuracy: 0.9573
+- F1: 0.9591
 
 ## Model description
 
@@ -44,21 +46,22 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 4
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step  | Validation Loss | F1     |
-|:-------------:|:-----:|:-----:|:---------------:|:------:|
-| 0.6425        | 1.0   | 3890  | 0.6213          | 0.7496 |
-| 0.6021        | 2.0   | 7780  | 0.5957          | 0.7436 |
-| 0.612         | 3.0   | 11670 | 0.5889          | 0.7477 |
+| Training Loss | Epoch | Step  | Validation Loss | Accuracy | F1     |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|
+| 0.0442        | 1.0   | 4238  | 0.0326          | 0.5051   | 0.6712 |
+| 0.0222        | 2.0   | 8476  | 0.0337          | 0.9391   | 0.9425 |
+| 0.0127        | 3.0   | 12714 | 0.0303          | 0.5756   | 0.7041 |
+| 0.0085        | 4.0   | 16952 | 0.0296          | 0.9573   | 0.9591 |
 
 
 ### Framework versions
 
-- Transformers 4.53.0
+- Transformers 4.53.2
 - Pytorch 2.6.0+cu124
 - Datasets 2.14.4
 - Tokenizers 0.21.2

config.json

@@ -7,8 +7,14 @@
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0"
+  },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0
+  },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
   "model_type": "bert",
@@ -16,9 +22,9 @@
   "num_hidden_layers": 12,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
-  "problem_type": "single_label_classification",
+  "problem_type": "regression",
   "torch_dtype": "float32",
-  "transformers_version": "4.53.0",
+  "transformers_version": "4.53.2",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 32000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5233dd37a95b0625d5f3d2e1f04fc226ea5d6d29e7fdad987d06aaa2df976f26
-size 442499064
+oid sha256:f9081ffb1c7ac8584e35a81d7b5b6bd1219af8ff7e931d98a5e7e3b2835df297
+size 442495988

tokenizer.json

@@ -7,9 +7,7 @@
     "stride": 0
   },
   "padding": {
-    "strategy": {
-      "Fixed": 512
-    },
+    "strategy": "BatchLongest",
     "direction": "Right",
     "pad_to_multiple_of": null,
     "pad_id": 0,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de41c0e9c863d94c431f6eae31c7a4c4d164738821f8e050c6fefa93c09bd408
+oid sha256:32a25b4c27cc71f7b1daf5746421d6ffdcd5a3066502fa6775acd7af2df1431e
 size 5304