End of training

README.md

@@ -1,5 +1,6 @@
 ---
-base_model: anton-l/wav2vec2-base-ft-common-language
+license: apache-2.0
+base_model: facebook/wav2vec2-base
 tags:
 - generated_from_trainer
 metrics:
@@ -14,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # AudioClassification
 
-This model is a fine-tuned version of [anton-l/wav2vec2-base-ft-common-language](https://huggingface.co/anton-l/wav2vec2-base-ft-common-language) on the None dataset.
+This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0348
+- Loss: 0.9903
 - Accuracy: 0.0
 
 ## Model description
@@ -45,17 +46,22 @@ The following hyperparameters were used during training:
 - 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
+- num_epochs: 10
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.6876        | 0.98  | 11   | 1.0517          | 0.0      |
-| 0.6753        | 1.96  | 22   | 1.0793          | 0.0      |
-| 0.6638        | 2.93  | 33   | 0.9722          | 0.0      |
-| 0.6627        | 4.0   | 45   | 1.0017          | 0.0      |
-| 0.6603        | 4.89  | 55   | 1.0348          | 0.0      |
+| 0.6783        | 0.98  | 11   | 1.0771          | 0.0      |
+| 0.6614        | 1.96  | 22   | 0.9514          | 0.0      |
+| 0.6628        | 2.93  | 33   | 0.9843          | 0.0      |
+| 0.6629        | 4.0   | 45   | 1.0408          | 0.0      |
+| 0.6583        | 4.98  | 56   | 1.0061          | 0.0      |
+| 0.6623        | 5.96  | 67   | 1.0227          | 0.0      |
+| 0.6613        | 6.93  | 78   | 1.0398          | 0.0      |
+| 0.6635        | 8.0   | 90   | 1.0085          | 0.0      |
+| 0.6577        | 8.98  | 101  | 0.9842          | 0.0      |
+| 0.6629        | 9.78  | 110  | 0.9903          | 0.0      |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "anton-l/wav2vec2-base-ft-common-language",
+  "_name_or_path": "facebook/wav2vec2-base",
   "activation_dropout": 0.0,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,
@@ -52,7 +52,6 @@
   "feat_proj_dropout": 0.1,
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
-  "finetuning_task": "audio-classification",
   "freeze_feat_extract_train": true,
   "hidden_act": "gelu",
   "hidden_dropout": 0.1,
@@ -68,7 +67,7 @@
     "uzbek": 0
   },
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.05,
+  "layerdrop": 0.0,
   "mask_channel_length": 10,
   "mask_channel_min_space": 1,
   "mask_channel_other": 0.0,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:582ef57a27ffa6ed455b986cb2e156179c72bee4df060785c8cdce164e20cd6a
+oid sha256:590a99a5755ba7529c930e56daad1201a0ef208bac93c6aec8e53e0a261f86fb
 size 378350950

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49adf150f8238474f2b72162d1ba9dd138d49388992cbfa6e90df27d5bc95f1c
+oid sha256:e5cc735e157f9801e030937ea4a1fc9896f4838e6cd8bd59ed4637b7d9d70045
 size 4536