Commit
•
501c2e5
1
Parent(s):
12fc299
Training in progress, epoch 1
Browse files- README.md +102 -0
- config.json +96 -96
- model.safetensors +1 -1
- preprocessor_config.json +9 -9
- training_args.bin +1 -1
README.md
ADDED
|
@@ -0,0 +1,102 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: ntu-spml/distilhubert
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- marsyas/gtzan
|
| 9 |
+
metrics:
|
| 10 |
+
- accuracy
|
| 11 |
+
model-index:
|
| 12 |
+
- name: distilhubert-finetuned-gtzan
|
| 13 |
+
results:
|
| 14 |
+
- task:
|
| 15 |
+
name: Audio Classification
|
| 16 |
+
type: audio-classification
|
| 17 |
+
dataset:
|
| 18 |
+
name: GTZAN
|
| 19 |
+
type: marsyas/gtzan
|
| 20 |
+
config: default
|
| 21 |
+
split: train
|
| 22 |
+
args: default
|
| 23 |
+
metrics:
|
| 24 |
+
- name: Accuracy
|
| 25 |
+
type: accuracy
|
| 26 |
+
value: 0.86
|
| 27 |
+
---
|
| 28 |
+
|
| 29 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 30 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 31 |
+
|
| 32 |
+
# distilhubert-finetuned-gtzan
|
| 33 |
+
|
| 34 |
+
This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the GTZAN dataset.
|
| 35 |
+
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 1.1749
|
| 37 |
+
- Accuracy: 0.86
|
| 38 |
+
|
| 39 |
+
## Model description
|
| 40 |
+
|
| 41 |
+
More information needed
|
| 42 |
+
|
| 43 |
+
## Intended uses & limitations
|
| 44 |
+
|
| 45 |
+
More information needed
|
| 46 |
+
|
| 47 |
+
## Training and evaluation data
|
| 48 |
+
|
| 49 |
+
More information needed
|
| 50 |
+
|
| 51 |
+
## Training procedure
|
| 52 |
+
|
| 53 |
+
### Training hyperparameters
|
| 54 |
+
|
| 55 |
+
The following hyperparameters were used during training:
|
| 56 |
+
- learning_rate: 5e-05
|
| 57 |
+
- train_batch_size: 4
|
| 58 |
+
- eval_batch_size: 4
|
| 59 |
+
- seed: 42
|
| 60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 61 |
+
- lr_scheduler_type: linear
|
| 62 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 63 |
+
- num_epochs: 25
|
| 64 |
+
- mixed_precision_training: Native AMP
|
| 65 |
+
|
| 66 |
+
### Training results
|
| 67 |
+
|
| 68 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 70 |
+
| 2.0517 | 1.0 | 225 | 2.0004 | 0.47 |
|
| 71 |
+
| 1.3283 | 2.0 | 450 | 1.3458 | 0.57 |
|
| 72 |
+
| 0.729 | 3.0 | 675 | 0.8563 | 0.76 |
|
| 73 |
+
| 0.4007 | 4.0 | 900 | 0.6748 | 0.8 |
|
| 74 |
+
| 0.3923 | 5.0 | 1125 | 0.7340 | 0.78 |
|
| 75 |
+
| 0.2193 | 6.0 | 1350 | 0.8712 | 0.76 |
|
| 76 |
+
| 0.2383 | 7.0 | 1575 | 0.7414 | 0.79 |
|
| 77 |
+
| 0.3 | 8.0 | 1800 | 0.7387 | 0.86 |
|
| 78 |
+
| 0.006 | 9.0 | 2025 | 0.9203 | 0.85 |
|
| 79 |
+
| 0.002 | 10.0 | 2250 | 0.8956 | 0.85 |
|
| 80 |
+
| 0.0014 | 11.0 | 2475 | 0.9831 | 0.86 |
|
| 81 |
+
| 0.001 | 12.0 | 2700 | 0.9406 | 0.86 |
|
| 82 |
+
| 0.0009 | 13.0 | 2925 | 1.0288 | 0.86 |
|
| 83 |
+
| 0.0007 | 14.0 | 3150 | 1.0172 | 0.86 |
|
| 84 |
+
| 0.0007 | 15.0 | 3375 | 0.9912 | 0.89 |
|
| 85 |
+
| 0.0005 | 16.0 | 3600 | 1.0282 | 0.86 |
|
| 86 |
+
| 0.0006 | 17.0 | 3825 | 1.3495 | 0.83 |
|
| 87 |
+
| 0.2453 | 18.0 | 4050 | 1.0340 | 0.87 |
|
| 88 |
+
| 0.0004 | 19.0 | 4275 | 1.1048 | 0.86 |
|
| 89 |
+
| 0.0004 | 20.0 | 4500 | 1.3051 | 0.85 |
|
| 90 |
+
| 0.0003 | 21.0 | 4725 | 1.2280 | 0.85 |
|
| 91 |
+
| 0.0003 | 22.0 | 4950 | 1.2530 | 0.85 |
|
| 92 |
+
| 0.0003 | 23.0 | 5175 | 1.1992 | 0.85 |
|
| 93 |
+
| 0.0003 | 24.0 | 5400 | 1.1881 | 0.85 |
|
| 94 |
+
| 0.0003 | 25.0 | 5625 | 1.1749 | 0.86 |
|
| 95 |
+
|
| 96 |
+
|
| 97 |
+
### Framework versions
|
| 98 |
+
|
| 99 |
+
- Transformers 4.44.2
|
| 100 |
+
- Pytorch 2.3.1
|
| 101 |
+
- Datasets 2.21.0
|
| 102 |
+
- Tokenizers 0.19.1
|
config.json
CHANGED
|
@@ -1,96 +1,96 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_name_or_path": "ntu-spml/distilhubert",
|
| 3 |
-
"activation_dropout": 0.1,
|
| 4 |
-
"apply_spec_augment": false,
|
| 5 |
-
"architectures": [
|
| 6 |
-
"HubertForSequenceClassification"
|
| 7 |
-
],
|
| 8 |
-
"attention_dropout": 0.1,
|
| 9 |
-
"bos_token_id": 1,
|
| 10 |
-
"classifier_proj_size": 256,
|
| 11 |
-
"conv_bias": false,
|
| 12 |
-
"conv_dim": [
|
| 13 |
-
512,
|
| 14 |
-
512,
|
| 15 |
-
512,
|
| 16 |
-
512,
|
| 17 |
-
512,
|
| 18 |
-
512,
|
| 19 |
-
512
|
| 20 |
-
],
|
| 21 |
-
"conv_kernel": [
|
| 22 |
-
10,
|
| 23 |
-
3,
|
| 24 |
-
3,
|
| 25 |
-
3,
|
| 26 |
-
3,
|
| 27 |
-
2,
|
| 28 |
-
2
|
| 29 |
-
],
|
| 30 |
-
"conv_stride": [
|
| 31 |
-
5,
|
| 32 |
-
2,
|
| 33 |
-
2,
|
| 34 |
-
2,
|
| 35 |
-
2,
|
| 36 |
-
2,
|
| 37 |
-
2
|
| 38 |
-
],
|
| 39 |
-
"ctc_loss_reduction": "sum",
|
| 40 |
-
"ctc_zero_infinity": false,
|
| 41 |
-
"do_stable_layer_norm": false,
|
| 42 |
-
"eos_token_id": 2,
|
| 43 |
-
"feat_extract_activation": "gelu",
|
| 44 |
-
"feat_extract_norm": "group",
|
| 45 |
-
"feat_proj_dropout": 0.0,
|
| 46 |
-
"feat_proj_layer_norm": false,
|
| 47 |
-
"final_dropout": 0.0,
|
| 48 |
-
"hidden_act": "gelu",
|
| 49 |
-
"hidden_dropout": 0.1,
|
| 50 |
-
"hidden_size": 768,
|
| 51 |
-
"id2label": {
|
| 52 |
-
"0": "blues",
|
| 53 |
-
"1": "classical",
|
| 54 |
-
"2": "country",
|
| 55 |
-
"3": "disco",
|
| 56 |
-
"4": "hiphop",
|
| 57 |
-
"5": "jazz",
|
| 58 |
-
"6": "metal",
|
| 59 |
-
"7": "pop",
|
| 60 |
-
"8": "reggae",
|
| 61 |
-
"9": "rock"
|
| 62 |
-
},
|
| 63 |
-
"initializer_range": 0.02,
|
| 64 |
-
"intermediate_size": 3072,
|
| 65 |
-
"label2id": {
|
| 66 |
-
"blues": "0",
|
| 67 |
-
"classical": "1",
|
| 68 |
-
"country": "2",
|
| 69 |
-
"disco": "3",
|
| 70 |
-
"hiphop": "4",
|
| 71 |
-
"jazz": "5",
|
| 72 |
-
"metal": "6",
|
| 73 |
-
"pop": "7",
|
| 74 |
-
"reggae": "8",
|
| 75 |
-
"rock": "9"
|
| 76 |
-
},
|
| 77 |
-
"layer_norm_eps": 1e-05,
|
| 78 |
-
"layerdrop": 0.0,
|
| 79 |
-
"mask_feature_length": 10,
|
| 80 |
-
"mask_feature_min_masks": 0,
|
| 81 |
-
"mask_feature_prob": 0.0,
|
| 82 |
-
"mask_time_length": 10,
|
| 83 |
-
"mask_time_min_masks": 2,
|
| 84 |
-
"mask_time_prob": 0.05,
|
| 85 |
-
"model_type": "hubert",
|
| 86 |
-
"num_attention_heads": 12,
|
| 87 |
-
"num_conv_pos_embedding_groups": 16,
|
| 88 |
-
"num_conv_pos_embeddings": 128,
|
| 89 |
-
"num_feat_extract_layers": 7,
|
| 90 |
-
"num_hidden_layers": 2,
|
| 91 |
-
"pad_token_id": 0,
|
| 92 |
-
"torch_dtype": "float32",
|
| 93 |
-
"transformers_version": "4.
|
| 94 |
-
"use_weighted_layer_sum": false,
|
| 95 |
-
"vocab_size": 32
|
| 96 |
-
}
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ntu-spml/distilhubert",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": false,
|
| 47 |
+
"final_dropout": 0.0,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 768,
|
| 51 |
+
"id2label": {
|
| 52 |
+
"0": "blues",
|
| 53 |
+
"1": "classical",
|
| 54 |
+
"2": "country",
|
| 55 |
+
"3": "disco",
|
| 56 |
+
"4": "hiphop",
|
| 57 |
+
"5": "jazz",
|
| 58 |
+
"6": "metal",
|
| 59 |
+
"7": "pop",
|
| 60 |
+
"8": "reggae",
|
| 61 |
+
"9": "rock"
|
| 62 |
+
},
|
| 63 |
+
"initializer_range": 0.02,
|
| 64 |
+
"intermediate_size": 3072,
|
| 65 |
+
"label2id": {
|
| 66 |
+
"blues": "0",
|
| 67 |
+
"classical": "1",
|
| 68 |
+
"country": "2",
|
| 69 |
+
"disco": "3",
|
| 70 |
+
"hiphop": "4",
|
| 71 |
+
"jazz": "5",
|
| 72 |
+
"metal": "6",
|
| 73 |
+
"pop": "7",
|
| 74 |
+
"reggae": "8",
|
| 75 |
+
"rock": "9"
|
| 76 |
+
},
|
| 77 |
+
"layer_norm_eps": 1e-05,
|
| 78 |
+
"layerdrop": 0.0,
|
| 79 |
+
"mask_feature_length": 10,
|
| 80 |
+
"mask_feature_min_masks": 0,
|
| 81 |
+
"mask_feature_prob": 0.0,
|
| 82 |
+
"mask_time_length": 10,
|
| 83 |
+
"mask_time_min_masks": 2,
|
| 84 |
+
"mask_time_prob": 0.05,
|
| 85 |
+
"model_type": "hubert",
|
| 86 |
+
"num_attention_heads": 12,
|
| 87 |
+
"num_conv_pos_embedding_groups": 16,
|
| 88 |
+
"num_conv_pos_embeddings": 128,
|
| 89 |
+
"num_feat_extract_layers": 7,
|
| 90 |
+
"num_hidden_layers": 2,
|
| 91 |
+
"pad_token_id": 0,
|
| 92 |
+
"torch_dtype": "float32",
|
| 93 |
+
"transformers_version": "4.44.2",
|
| 94 |
+
"use_weighted_layer_sum": false,
|
| 95 |
+
"vocab_size": 32
|
| 96 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 94771728
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d624d0e27bcd2f949460bb8992b8e4a6267d723c784c9a27714d942eec60b06
|
| 3 |
size 94771728
|
preprocessor_config.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
-
{
|
| 2 |
-
"do_normalize": true,
|
| 3 |
-
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
-
"feature_size": 1,
|
| 5 |
-
"padding_side": "right",
|
| 6 |
-
"padding_value": 0,
|
| 7 |
-
"return_attention_mask": true,
|
| 8 |
-
"sampling_rate": 16000
|
| 9 |
-
}
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5176
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebe1e48b34c662b94395ef1115e514706f0ebe3a462213cba751f336a9e1d94c
|
| 3 |
size 5176
|