End of training

README.md

@@ -1,39 +1,19 @@
 ---
 license: apache-2.0
-base_model: facebook/wav2vec2-xls-r-300m
+base_model: Amrrs/wav2vec2-large-xlsr-53-tamil
 tags:
 - generated_from_trainer
-datasets:
-- common_voice_12_0
-metrics:
-- wer
 model-index:
-- name: wav2vec2-large-xls-r-300m-tamil2-colab
-  results:
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: common_voice_12_0
-      type: common_voice_12_0
-      config: ta
-      split: test[:5%]
-      args: ta
-    metrics:
-    - name: Wer
-      type: wer
-      value: 1.0
+- name: wav2vec2-large-xls-r-300m-Tamil2-colab
+  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-large-xls-r-300m-tamil2-colab
+# wav2vec2-large-xls-r-300m-Tamil2-colab
 
-This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice_12_0 dataset.
-It achieves the following results on the evaluation set:
-- Loss: 3.1835
-- Wer: 1.0
+This model is a fine-tuned version of [Amrrs/wav2vec2-large-xlsr-53-tamil](https://huggingface.co/Amrrs/wav2vec2-large-xlsr-53-tamil) on the None dataset.
 
 ## Model description
 
@@ -53,29 +33,19 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0003
-- train_batch_size: 16
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 32
+- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 50
-- num_epochs: 5
+- num_epochs: 2
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Wer |
-|:-------------:|:------:|:----:|:---------------:|:---:|
-| 10.7536       | 0.5970 | 40   | 4.9828          | 1.0 |
-| 3.5054        | 1.1940 | 80   | 3.3975          | 1.0 |
-| 3.2253        | 1.7910 | 120  | 3.3687          | 1.0 |
-| 3.185         | 2.3881 | 160  | 3.2591          | 1.0 |
-| 3.1631        | 2.9851 | 200  | 3.2295          | 1.0 |
-| 3.1451        | 3.5821 | 240  | 3.2064          | 1.0 |
-| 3.1281        | 4.1791 | 280  | 3.1945          | 1.0 |
-| 3.1231        | 4.7761 | 320  | 3.1835          | 1.0 |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "_name_or_path": "Amrrs/wav2vec2-large-xlsr-53-tamil",
   "activation_dropout": 0.0,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,
@@ -9,10 +9,10 @@
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.2,
+  "attention_dropout": 0.1,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
-  "codevector_dim": 768,
+  "codevector_dim": 256,
   "contrastive_logits_temperature": 0.1,
   "conv_bias": true,
   "conv_dim": [
@@ -54,18 +54,26 @@
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.0,
+  "hidden_dropout": 0.1,
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.0,
+  "layerdrop": 0.1,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
   "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_min_space": 1,
+  "mask_time_other": 0.0,
   "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
   "model_type": "wav2vec2",
   "num_adapter_layers": 3,
   "num_attention_heads": 16,
@@ -77,8 +85,8 @@
   "num_hidden_layers": 24,
   "num_negatives": 100,
   "output_hidden_size": 1024,
-  "pad_token_id": 57,
-  "proj_codevector_dim": 768,
+  "pad_token_id": 51,
+  "proj_codevector_dim": 256,
   "tdnn_dilation": [
     1,
     2,
@@ -103,6 +111,6 @@
   "torch_dtype": "float32",
   "transformers_version": "4.41.2",
   "use_weighted_layer_sum": false,
-  "vocab_size": 60,
+  "vocab_size": 52,
   "xvector_output_dim": 512
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c6a599274c98477615bb1dafaefbbe38828af35311cecfd21151fcd68a2083c6
-size 1262053480
+oid sha256:c5a6b1f7a4942d8393e3d748bfee87ee24eb05f9277d78a45186f1775c28e500
+size 1262020680

runs/Jun26_07-17-03_2e29ec53aaf1/events.out.tfevents.1719386266.2e29ec53aaf1.150.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2972e271e218a02425e5bf671ddf22e49a0fd9570845e4e2414e602566923eea
+size 6846

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32cdd5c4dfc6e62b4b5554d30df58da75c972eafbc01065ea779d9f14d6b65fd
+oid sha256:157b23262ec98eac79179cd488d689286eefe54e696340d9b779cbbf4b03b208
 size 5112