End of training

README.md

@@ -0,0 +1,74 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: team-lucid/hubert-base-korean
+tags:
+- generated_from_trainer
+model-index:
+- name: for_test13
+  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. -->
+
+# for_test13
+
+This model is a fine-tuned version of [team-lucid/hubert-base-korean](https://huggingface.co/team-lucid/hubert-base-korean) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 9.8047
+- Per: 2.5339
+- Learning Rate: 0.0000
+
+## 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: 0.0001
+- train_batch_size: 32
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 20
+- num_epochs: 10
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Per    | Rate   |
+|:-------------:|:------:|:----:|:---------------:|:------:|:------:|
+| 10.7575       | 0.8547 | 50   | 11.3211         | 1.3831 | 0.0001 |
+| 6.5834        | 1.7094 | 100  | 10.3838         | 2.6287 | 0.0001 |
+| 5.7953        | 2.5641 | 150  | 10.0381         | 2.7805 | 0.0001 |
+| 5.4816        | 3.4188 | 200  | 9.9584          | 2.7593 | 0.0001 |
+| 5.4448        | 4.2735 | 250  | 9.9493          | 2.7246 | 0.0001 |
+| 5.1187        | 5.1282 | 300  | 9.9573          | 2.6414 | 0.0001 |
+| 4.5692        | 5.9829 | 350  | 9.9676          | 2.5711 | 0.0000 |
+| 5.1398        | 6.8376 | 400  | 9.8478          | 2.5756 | 0.0000 |
+| 4.7656        | 7.6923 | 450  | 9.8491          | 2.5501 | 0.0000 |
+| 5.0038        | 8.5470 | 500  | 9.8228          | 2.5269 | 0.0000 |
+| 5.3115        | 9.4017 | 550  | 9.8047          | 2.5339 | 0.0000 |
+
+
+### Framework versions
+
+- Transformers 4.46.2
+- Pytorch 2.5.1+cu121
+- Datasets 3.1.0
+- Tokenizers 0.20.3

config.json

@@ -0,0 +1,80 @@
+{
+  "_name_or_path": "team-lucid/hubert-base-korean",
+  "activation_dropout": 0.05,
+  "apply_spec_augment": true,
+  "architectures": [
+    "HubertForCTC"
+  ],
+  "attention_dropout": 0.01,
+  "auto_map": {
+    "FlaxAutoModel": "team-lucid/hubert-base-korean--modeling_flax_hubert.FlaxHubertModel"
+  },
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "conv_bias": false,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": true,
+  "do_stable_layer_norm": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.05,
+  "feat_proj_layer_norm": true,
+  "final_dropout": 0.1,
+  "final_proj_size": 256,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.05,
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.1,
+  "model_type": "hubert",
+  "num_attention_heads": 12,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 12,
+  "pad_token_id": 43,
+  "tokenizer_class": "Wav2Vec2CTCTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.46.2",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 44
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11fedae78d85046497f3ce8a233a9904ce954e7e6eacd82c59034e120bace9f8
+size 377647624

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "cache_dir": "/content/drive/MyDrive/here",
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05e8c0e0ddea3f29f722a59cbf32ca7d2cdcfd90d776f426cefb721f35733a36
+size 5240