Model save

README.md

@@ -0,0 +1,83 @@
+---
+license: apache-2.0
+base_model: Bisher/wav2vec2_ASV_deepfake_audio_detection_DF_finetune_frozen
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+- precision
+- recall
+- f1
+model-index:
+- name: wav2vec2_ASV_deepfake_audio_detection_DF_finetune_frozen_chngd_classifier
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# wav2vec2_ASV_deepfake_audio_detection_DF_finetune_frozen_chngd_classifier
+
+This model is a fine-tuned version of [Bisher/wav2vec2_ASV_deepfake_audio_detection_DF_finetune_frozen](https://huggingface.co/Bisher/wav2vec2_ASV_deepfake_audio_detection_DF_finetune_frozen) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4087
+- Accuracy: 0.9167
+- Precision: 0.9180
+- Recall: 0.9167
+- F1: 0.8908
+- Tp: 375
+- Tn: 17886
+- Fn: 1632
+- Fp: 27
+- Eer: 0.1674
+- Min Tdcf: 0.0312
+- Auc Roc: 0.8827
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0005
+- train_batch_size: 64
+- eval_batch_size: 64
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 256
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 5
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy | Precision | Recall | F1     | Tp   | Tn    | Fn   | Fp  | Eer    | Min Tdcf | Auc Roc |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:----:|:-----:|:----:|:---:|:------:|:--------:|:-------:|
+| 0.7199        | 0.0103 | 5    | 0.6815          | 0.9488   | 0.9472    | 0.9488 | 0.9478 | 1413 | 17488 | 594  | 425 | 0.1520 | 0.0329   | 0.8893  |
+| 0.6841        | 0.0206 | 10   | 0.6499          | 0.9304   | 0.9301    | 0.9304 | 0.9154 | 675  | 17859 | 1332 | 54  | 0.2820 | 0.0325   | 0.7611  |
+| 0.6394        | 0.0309 | 15   | 0.5997          | 0.9079   | 0.9088    | 0.9079 | 0.8726 | 189  | 17896 | 1818 | 17  | 0.1166 | 0.0323   | 0.9573  |
+| 0.5966        | 0.0413 | 20   | 0.5522          | 0.9020   | 0.9083    | 0.9020 | 0.8583 | 57   | 17911 | 1950 | 2   | 0.0837 | 0.0330   | 0.9745  |
+| 0.5565        | 0.0516 | 25   | 0.5057          | 0.9030   | 0.9067    | 0.9030 | 0.8609 | 80   | 17908 | 1927 | 5   | 0.0872 | 0.0333   | 0.9712  |
+| 0.5236        | 0.0619 | 30   | 0.4542          | 0.9124   | 0.9141    | 0.9124 | 0.8822 | 284  | 17892 | 1723 | 21  | 0.0876 | 0.0315   | 0.9648  |
+| 0.4739        | 0.0722 | 35   | 0.4087          | 0.9167   | 0.9180    | 0.9167 | 0.8908 | 375  | 17886 | 1632 | 27  | 0.1674 | 0.0312   | 0.8827  |
+
+
+### Framework versions
+
+- Transformers 4.44.0
+- Pytorch 2.4.0
+- Datasets 2.21.0
+- Tokenizers 0.19.1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f1d93f241dbf41582b888d3e93a392226a0affb967276591a7b383cff1f88d3
+oid sha256:cd45d5195eecafa09f6a65213a143f9fb79e53e6b29d41590bce69e0a092ac1e
 size 378449016