dmusingu commited on
Commit
7f2b137
·
verified ·
1 Parent(s): 9df3209

dmusingu/muk-english-digits-classification

Browse files
Files changed (5) hide show
  1. README.md +111 -0
  2. config.json +141 -0
  3. model.safetensors +3 -0
  4. preprocessor_config.json +9 -0
  5. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,111 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: facebook/wav2vec2-base
5
+ tags:
6
+ - generated_from_trainer
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: lg-digits-classification
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # lg-digits-classification
18
+
19
+ This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 2.5405
22
+ - Accuracy: 0.6087
23
+
24
+ ## Model description
25
+
26
+ More information needed
27
+
28
+ ## Intended uses & limitations
29
+
30
+ More information needed
31
+
32
+ ## Training and evaluation data
33
+
34
+ More information needed
35
+
36
+ ## Training procedure
37
+
38
+ ### Training hyperparameters
39
+
40
+ The following hyperparameters were used during training:
41
+ - learning_rate: 3e-05
42
+ - train_batch_size: 4
43
+ - eval_batch_size: 1
44
+ - seed: 42
45
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_ratio: 0.1
48
+ - num_epochs: 50
49
+
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
53
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
54
+ | 1.9211 | 1.0 | 47 | 2.0718 | 0.2174 |
55
+ | 1.9034 | 2.0 | 94 | 2.0263 | 0.2174 |
56
+ | 1.8375 | 3.0 | 141 | 2.1374 | 0.2609 |
57
+ | 1.7839 | 4.0 | 188 | 2.0091 | 0.3043 |
58
+ | 1.6964 | 5.0 | 235 | 2.0135 | 0.2609 |
59
+ | 1.6485 | 6.0 | 282 | 1.9707 | 0.4348 |
60
+ | 1.5452 | 7.0 | 329 | 1.8629 | 0.3043 |
61
+ | 1.4662 | 8.0 | 376 | 1.8037 | 0.4348 |
62
+ | 1.3512 | 9.0 | 423 | 1.6884 | 0.5217 |
63
+ | 1.2567 | 10.0 | 470 | 1.7180 | 0.5652 |
64
+ | 1.1825 | 11.0 | 517 | 1.5855 | 0.5652 |
65
+ | 1.0535 | 12.0 | 564 | 1.6781 | 0.4783 |
66
+ | 0.9519 | 13.0 | 611 | 1.5681 | 0.5652 |
67
+ | 0.9136 | 14.0 | 658 | 1.4914 | 0.6087 |
68
+ | 0.7924 | 15.0 | 705 | 1.7370 | 0.5652 |
69
+ | 0.6775 | 16.0 | 752 | 1.6007 | 0.5652 |
70
+ | 0.7597 | 17.0 | 799 | 1.5242 | 0.6087 |
71
+ | 0.5731 | 18.0 | 846 | 1.7712 | 0.6087 |
72
+ | 0.5235 | 19.0 | 893 | 1.5915 | 0.5652 |
73
+ | 0.5463 | 20.0 | 940 | 1.7188 | 0.5652 |
74
+ | 0.4434 | 21.0 | 987 | 1.8435 | 0.5652 |
75
+ | 0.3921 | 22.0 | 1034 | 1.8180 | 0.6522 |
76
+ | 0.4408 | 23.0 | 1081 | 1.9594 | 0.5652 |
77
+ | 0.3323 | 24.0 | 1128 | 1.9446 | 0.5652 |
78
+ | 0.2806 | 25.0 | 1175 | 1.9466 | 0.5652 |
79
+ | 0.2228 | 26.0 | 1222 | 2.1430 | 0.5652 |
80
+ | 0.3144 | 27.0 | 1269 | 2.2630 | 0.5652 |
81
+ | 0.2841 | 28.0 | 1316 | 2.1039 | 0.5652 |
82
+ | 0.2392 | 29.0 | 1363 | 2.1887 | 0.6087 |
83
+ | 0.1964 | 30.0 | 1410 | 2.1597 | 0.6522 |
84
+ | 0.2014 | 31.0 | 1457 | 2.2470 | 0.6087 |
85
+ | 0.194 | 32.0 | 1504 | 2.3058 | 0.5652 |
86
+ | 0.202 | 33.0 | 1551 | 2.4327 | 0.5652 |
87
+ | 0.1968 | 34.0 | 1598 | 2.4571 | 0.5652 |
88
+ | 0.1675 | 35.0 | 1645 | 2.4031 | 0.6087 |
89
+ | 0.2261 | 36.0 | 1692 | 2.4036 | 0.6087 |
90
+ | 0.279 | 37.0 | 1739 | 2.4073 | 0.6087 |
91
+ | 0.2378 | 38.0 | 1786 | 2.4481 | 0.6087 |
92
+ | 0.2414 | 39.0 | 1833 | 2.4839 | 0.6087 |
93
+ | 0.189 | 40.0 | 1880 | 2.5730 | 0.5652 |
94
+ | 0.2018 | 41.0 | 1927 | 2.6159 | 0.5652 |
95
+ | 0.2418 | 42.0 | 1974 | 2.4725 | 0.5652 |
96
+ | 0.1489 | 43.0 | 2021 | 2.4596 | 0.5652 |
97
+ | 0.2277 | 44.0 | 2068 | 2.4913 | 0.6087 |
98
+ | 0.2313 | 45.0 | 2115 | 2.4922 | 0.6087 |
99
+ | 0.2333 | 46.0 | 2162 | 2.4951 | 0.6087 |
100
+ | 0.2201 | 47.0 | 2209 | 2.5213 | 0.6087 |
101
+ | 0.2116 | 48.0 | 2256 | 2.5164 | 0.6087 |
102
+ | 0.2122 | 49.0 | 2303 | 2.5424 | 0.6087 |
103
+ | 0.1965 | 50.0 | 2350 | 2.5405 | 0.6087 |
104
+
105
+
106
+ ### Framework versions
107
+
108
+ - Transformers 4.56.2
109
+ - Pytorch 2.8.0+cu128
110
+ - Datasets 3.6.0
111
+ - Tokenizers 0.22.1
config.json ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.0,
3
+ "adapter_attn_dim": null,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "Wav2Vec2ForSequenceClassification"
10
+ ],
11
+ "attention_dropout": 0.1,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 256,
15
+ "contrastive_logits_temperature": 0.1,
16
+ "conv_bias": false,
17
+ "conv_dim": [
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512
25
+ ],
26
+ "conv_kernel": [
27
+ 10,
28
+ 3,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 2,
33
+ 2
34
+ ],
35
+ "conv_stride": [
36
+ 5,
37
+ 2,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2
43
+ ],
44
+ "ctc_loss_reduction": "sum",
45
+ "ctc_zero_infinity": false,
46
+ "diversity_loss_weight": 0.1,
47
+ "do_stable_layer_norm": false,
48
+ "dtype": "float32",
49
+ "eos_token_id": 2,
50
+ "feat_extract_activation": "gelu",
51
+ "feat_extract_norm": "group",
52
+ "feat_proj_dropout": 0.1,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.0,
55
+ "freeze_feat_extract_train": true,
56
+ "hidden_act": "gelu",
57
+ "hidden_dropout": 0.1,
58
+ "hidden_size": 768,
59
+ "id2label": {
60
+ "0": "zero",
61
+ "1": "one",
62
+ "2": "two",
63
+ "3": "three",
64
+ "4": "four",
65
+ "5": "five",
66
+ "6": "six",
67
+ "7": "seven",
68
+ "8": "eight",
69
+ "9": "nine"
70
+ },
71
+ "initializer_range": 0.02,
72
+ "intermediate_size": 3072,
73
+ "label2id": {
74
+ "eight": 8,
75
+ "five": 5,
76
+ "four": 4,
77
+ "nine": 9,
78
+ "one": 1,
79
+ "seven": 7,
80
+ "six": 6,
81
+ "three": 3,
82
+ "two": 2,
83
+ "zero": 0
84
+ },
85
+ "layer_norm_eps": 1e-05,
86
+ "layerdrop": 0.0,
87
+ "mask_channel_length": 10,
88
+ "mask_channel_min_space": 1,
89
+ "mask_channel_other": 0.0,
90
+ "mask_channel_prob": 0.0,
91
+ "mask_channel_selection": "static",
92
+ "mask_feature_length": 10,
93
+ "mask_feature_min_masks": 0,
94
+ "mask_feature_prob": 0.0,
95
+ "mask_time_length": 10,
96
+ "mask_time_min_masks": 2,
97
+ "mask_time_min_space": 1,
98
+ "mask_time_other": 0.0,
99
+ "mask_time_prob": 0.05,
100
+ "mask_time_selection": "static",
101
+ "model_type": "wav2vec2",
102
+ "no_mask_channel_overlap": false,
103
+ "no_mask_time_overlap": false,
104
+ "num_adapter_layers": 3,
105
+ "num_attention_heads": 12,
106
+ "num_codevector_groups": 2,
107
+ "num_codevectors_per_group": 320,
108
+ "num_conv_pos_embedding_groups": 16,
109
+ "num_conv_pos_embeddings": 128,
110
+ "num_feat_extract_layers": 7,
111
+ "num_hidden_layers": 12,
112
+ "num_negatives": 100,
113
+ "output_hidden_size": 768,
114
+ "pad_token_id": 0,
115
+ "proj_codevector_dim": 256,
116
+ "tdnn_dilation": [
117
+ 1,
118
+ 2,
119
+ 3,
120
+ 1,
121
+ 1
122
+ ],
123
+ "tdnn_dim": [
124
+ 512,
125
+ 512,
126
+ 512,
127
+ 512,
128
+ 1500
129
+ ],
130
+ "tdnn_kernel": [
131
+ 5,
132
+ 3,
133
+ 3,
134
+ 1,
135
+ 1
136
+ ],
137
+ "transformers_version": "4.56.2",
138
+ "use_weighted_layer_sum": false,
139
+ "vocab_size": 32,
140
+ "xvector_output_dim": 512
141
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f62cf476cce1da8d2e6a0d9cbe72d8b151ef2b20f96ed5cfd8fb57d62e6c785e
3
+ size 378310592
preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": false,
8
+ "sampling_rate": 16000
9
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f48da41a05b88cfd62cc374d55cee58f8890ea7654db66ea47fa00b9f8def07a
3
+ size 5841