End of training

README.md

@@ -1,7 +1,7 @@
 ---
 library_name: transformers
-license: cc-by-nc-4.0
-base_model: facebook/mms-1b-all
+license: apache-2.0
+base_model: facebook/wav2vec2-xls-r-300m
 tags:
 - generated_from_trainer
 metrics:
@@ -16,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # training
 
-This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the None dataset.
+This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0874
-- Wer: 0.8986
+- Loss: 2.0763
+- Wer: 0.9551
 
 ## Model description
 
@@ -38,39 +38,34 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0001
+- learning_rate: 0.0003
 - train_batch_size: 4
-- eval_batch_size: 8
+- eval_batch_size: 4
 - seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 8
 - 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: 500
-- num_epochs: 10
+- num_epochs: 25
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Wer    |
-|:-------------:|:------:|:----:|:---------------:|:------:|
-| 5.9632        | 0.4926 | 500  | 1.0587          | 0.7186 |
-| 1.2221        | 0.9852 | 1000 | 0.9062          | 0.6633 |
-| 1.0028        | 1.4778 | 1500 | 0.8629          | 0.6419 |
-| 1.1607        | 1.9704 | 2000 | 0.8534          | 0.6337 |
-| 1.0608        | 2.4631 | 2500 | 1.0133          | 0.7632 |
-| 1.3117        | 2.9557 | 3000 | 1.0389          | 0.7365 |
-| 1.3259        | 3.4483 | 3500 | 1.0874          | 0.8991 |
-| 1.2260        | 3.9409 | 4000 | 1.0875          | 0.8983 |
-| 1.3359        | 4.4335 | 4500 | 1.0875          | 0.8989 |
-| 1.2309        | 4.9261 | 5000 | 1.0875          | 0.8979 |
-| 1.2623        | 5.4187 | 5500 | 1.0875          | 0.8985 |
-| 1.2284        | 5.9113 | 6000 | 1.0875          | 0.8982 |
-| 1.3132        | 6.4039 | 6500 | 1.0875          | 0.8985 |
-| 1.3229        | 6.8966 | 7000 | 1.0874          | 0.8986 |
+| Training Loss | Epoch   | Step | Validation Loss | Wer    |
+|:-------------:|:-------:|:----:|:---------------:|:------:|
+| 3.6154        | 3.3557  | 500  | 3.6223          | 1.0    |
+| 3.6729        | 6.7114  | 1000 | 3.5869          | 1.0    |
+| 3.6718        | 10.0671 | 1500 | 3.5824          | 1.0    |
+| 3.7365        | 13.4228 | 2000 | 3.5829          | 1.0    |
+| 3.5897        | 16.7785 | 2500 | 3.5750          | 1.0    |
+| 3.5689        | 20.1342 | 3000 | 3.2325          | 1.0    |
+| 2.4663        | 23.4899 | 3500 | 2.0763          | 0.9551 |
 
 
 ### Framework versions
 
-- Transformers 5.0.0.dev0
-- Pytorch 2.5.1+cu121
+- Transformers 4.57.3
+- Pytorch 2.6.0+cu124
 - Datasets 3.6.0
 - Tokenizers 0.22.1

config.json

@@ -1,6 +1,6 @@
 {
-  "activation_dropout": 0.05,
-  "adapter_attn_dim": 16,
+  "activation_dropout": 0.0,
+  "adapter_attn_dim": null,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,
   "add_adapter": false,
@@ -8,10 +8,10 @@
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.0,
+  "attention_dropout": 0.1,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
-  "codevector_dim": 1024,
+  "codevector_dim": 768,
   "contrastive_logits_temperature": 0.1,
   "conv_bias": true,
   "conv_dim": [
@@ -50,22 +50,23 @@
   "feat_extract_activation": "gelu",
   "feat_extract_dropout": 0.0,
   "feat_extract_norm": "layer",
-  "feat_proj_dropout": 0.0,
+  "feat_proj_dropout": 0.1,
   "feat_quantizer_dropout": 0.0,
-  "final_dropout": 0.05,
+  "final_dropout": 0.0,
+  "gradient_checkpointing": false,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.0,
-  "hidden_size": 1280,
+  "hidden_dropout": 0.1,
+  "hidden_size": 1024,
   "initializer_range": 0.02,
-  "intermediate_size": 5120,
+  "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.0,
+  "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": 0.05,
+  "mask_time_prob": 0.075,
   "model_type": "wav2vec2",
   "num_adapter_layers": 3,
   "num_attention_heads": 16,
@@ -74,11 +75,11 @@
   "num_conv_pos_embedding_groups": 16,
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
-  "num_hidden_layers": 48,
+  "num_hidden_layers": 24,
   "num_negatives": 100,
-  "output_hidden_size": 1280,
-  "pad_token_id": 73,
-  "proj_codevector_dim": 1024,
+  "output_hidden_size": 1024,
+  "pad_token_id": 0,
+  "proj_codevector_dim": 768,
   "tdnn_dilation": [
     1,
     2,
@@ -100,9 +101,8 @@
     1,
     1
   ],
-  "transformers_version": "5.0.0.dev0",
-  "use_cache": false,
+  "transformers_version": "4.57.3",
   "use_weighted_layer_sum": false,
-  "vocab_size": 76,
+  "vocab_size": 101,
   "xvector_output_dim": 512
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c24975998840862e3f61a01f98ebc32afe3591229979b25294fa9c5c33e1bdb
-size 3859121448
+oid sha256:d7b961a2ff48bf57947ed7c6742a6be0934d725f9b416ea5855291942677f157
+size 1262221580

preprocessor_config.json

@@ -5,6 +5,6 @@
   "padding_side": "right",
   "padding_value": 0.0,
   "processor_class": "Wav2Vec2Processor",
-  "return_attention_mask": true,
+  "return_attention_mask": false,
   "sampling_rate": 16000
 }

runs/Dec21_07-13-01_Hong/events.out.tfevents.1766275982.Hong.32364.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17d4fc86d6631e92e0687defdb2a9d1d6705b366a5679628e06e4e994731759e
+size 16807

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4dd97ab74010796555cc2b4db00f8eb2bdacb0bad31c58579c0fa23f46bd7c3
-size 4792
+oid sha256:27ec27fc626e80c861fd309a025d0fc91a401021154294796cf7d74cfd5a6fd4
+size 5432