dmusingu commited on
Commit
ac91b51
·
verified ·
1 Parent(s): e0a1442

dmusingu/muk-luganda-digits-classification

Browse files
Files changed (5) hide show
  1. README.md +136 -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 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: muk-luganda-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
+ # muk-luganda-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: 5.1901
22
+ - Accuracy: 0.2414
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: 75
49
+
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
53
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
54
+ | 2.304 | 1.0 | 58 | 2.3169 | 0.1034 |
55
+ | 2.2989 | 2.0 | 116 | 2.3166 | 0.0690 |
56
+ | 2.2921 | 3.0 | 174 | 2.3167 | 0.1379 |
57
+ | 2.2843 | 4.0 | 232 | 2.3237 | 0.1034 |
58
+ | 2.2657 | 5.0 | 290 | 2.3195 | 0.0690 |
59
+ | 2.2519 | 6.0 | 348 | 2.2997 | 0.1034 |
60
+ | 2.2187 | 7.0 | 406 | 2.3050 | 0.1034 |
61
+ | 2.15 | 8.0 | 464 | 2.2577 | 0.1034 |
62
+ | 2.0555 | 9.0 | 522 | 2.2375 | 0.1379 |
63
+ | 1.9332 | 10.0 | 580 | 2.2008 | 0.2414 |
64
+ | 1.8073 | 11.0 | 638 | 2.1950 | 0.1724 |
65
+ | 1.6742 | 12.0 | 696 | 2.2654 | 0.1724 |
66
+ | 1.5511 | 13.0 | 754 | 2.1905 | 0.1724 |
67
+ | 1.4453 | 14.0 | 812 | 2.2018 | 0.2069 |
68
+ | 1.3227 | 15.0 | 870 | 2.1572 | 0.2759 |
69
+ | 1.2019 | 16.0 | 928 | 2.4037 | 0.2414 |
70
+ | 1.1151 | 17.0 | 986 | 2.4147 | 0.3103 |
71
+ | 0.9743 | 18.0 | 1044 | 2.3563 | 0.2759 |
72
+ | 0.9336 | 19.0 | 1102 | 2.2765 | 0.3103 |
73
+ | 0.852 | 20.0 | 1160 | 2.3266 | 0.3448 |
74
+ | 0.7096 | 21.0 | 1218 | 2.2129 | 0.4138 |
75
+ | 0.6183 | 22.0 | 1276 | 2.4103 | 0.3103 |
76
+ | 0.6018 | 23.0 | 1334 | 2.8189 | 0.2759 |
77
+ | 0.5155 | 24.0 | 1392 | 2.9791 | 0.2759 |
78
+ | 0.4233 | 25.0 | 1450 | 2.9289 | 0.3103 |
79
+ | 0.4213 | 26.0 | 1508 | 3.2987 | 0.2759 |
80
+ | 0.4613 | 27.0 | 1566 | 3.1643 | 0.3103 |
81
+ | 0.4414 | 28.0 | 1624 | 2.9910 | 0.3448 |
82
+ | 0.4033 | 29.0 | 1682 | 3.2960 | 0.3103 |
83
+ | 0.3761 | 30.0 | 1740 | 3.5232 | 0.2759 |
84
+ | 0.2886 | 31.0 | 1798 | 3.9775 | 0.2414 |
85
+ | 0.3646 | 32.0 | 1856 | 3.4826 | 0.3103 |
86
+ | 0.3371 | 33.0 | 1914 | 3.5032 | 0.3793 |
87
+ | 0.2169 | 34.0 | 1972 | 3.4125 | 0.3448 |
88
+ | 0.2906 | 35.0 | 2030 | 3.3535 | 0.3793 |
89
+ | 0.414 | 36.0 | 2088 | 3.7734 | 0.2069 |
90
+ | 0.2058 | 37.0 | 2146 | 3.7419 | 0.2759 |
91
+ | 0.188 | 38.0 | 2204 | 3.7832 | 0.3103 |
92
+ | 0.2209 | 39.0 | 2262 | 3.6297 | 0.2414 |
93
+ | 0.2403 | 40.0 | 2320 | 3.7870 | 0.3103 |
94
+ | 0.2483 | 41.0 | 2378 | 3.9331 | 0.3448 |
95
+ | 0.2409 | 42.0 | 2436 | 4.1382 | 0.3103 |
96
+ | 0.2415 | 43.0 | 2494 | 4.1796 | 0.2759 |
97
+ | 0.1496 | 44.0 | 2552 | 4.1671 | 0.3448 |
98
+ | 0.1874 | 45.0 | 2610 | 4.5372 | 0.2069 |
99
+ | 0.2092 | 46.0 | 2668 | 4.8994 | 0.2414 |
100
+ | 0.2016 | 47.0 | 2726 | 4.4871 | 0.2414 |
101
+ | 0.259 | 48.0 | 2784 | 4.1237 | 0.2759 |
102
+ | 0.2078 | 49.0 | 2842 | 4.9369 | 0.2414 |
103
+ | 0.1413 | 50.0 | 2900 | 4.9249 | 0.2414 |
104
+ | 0.1453 | 51.0 | 2958 | 5.1858 | 0.2414 |
105
+ | 0.1963 | 52.0 | 3016 | 3.9934 | 0.3103 |
106
+ | 0.2249 | 53.0 | 3074 | 4.2996 | 0.3103 |
107
+ | 0.2058 | 54.0 | 3132 | 3.7626 | 0.3448 |
108
+ | 0.2483 | 55.0 | 3190 | 4.6947 | 0.2759 |
109
+ | 0.209 | 56.0 | 3248 | 4.0701 | 0.3793 |
110
+ | 0.144 | 57.0 | 3306 | 4.5588 | 0.2414 |
111
+ | 0.2093 | 58.0 | 3364 | 4.9072 | 0.3103 |
112
+ | 0.143 | 59.0 | 3422 | 5.1035 | 0.2414 |
113
+ | 0.2102 | 60.0 | 3480 | 5.3840 | 0.2069 |
114
+ | 0.1507 | 61.0 | 3538 | 5.3511 | 0.2414 |
115
+ | 0.1826 | 62.0 | 3596 | 5.0179 | 0.2069 |
116
+ | 0.1407 | 63.0 | 3654 | 5.2103 | 0.1724 |
117
+ | 0.1395 | 64.0 | 3712 | 5.4675 | 0.2069 |
118
+ | 0.1683 | 65.0 | 3770 | 5.3124 | 0.2414 |
119
+ | 0.1761 | 66.0 | 3828 | 4.8748 | 0.2414 |
120
+ | 0.1868 | 67.0 | 3886 | 4.9547 | 0.2414 |
121
+ | 0.2524 | 68.0 | 3944 | 4.9069 | 0.2759 |
122
+ | 0.1367 | 69.0 | 4002 | 4.8562 | 0.2414 |
123
+ | 0.1368 | 70.0 | 4060 | 4.9808 | 0.2414 |
124
+ | 0.1841 | 71.0 | 4118 | 5.1471 | 0.2414 |
125
+ | 0.1658 | 72.0 | 4176 | 5.1741 | 0.2414 |
126
+ | 0.1553 | 73.0 | 4234 | 5.2040 | 0.2414 |
127
+ | 0.1625 | 74.0 | 4292 | 5.1895 | 0.2414 |
128
+ | 0.1823 | 75.0 | 4350 | 5.1901 | 0.2414 |
129
+
130
+
131
+ ### Framework versions
132
+
133
+ - Transformers 4.56.2
134
+ - Pytorch 2.8.0+cu128
135
+ - Datasets 3.6.0
136
+ - 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": "notti",
61
+ "1": "emu",
62
+ "2": "bbiri",
63
+ "3": "ssatu",
64
+ "4": "nnya",
65
+ "5": "ttaano",
66
+ "6": "mukaaga",
67
+ "7": "musanvu",
68
+ "8": "munaana",
69
+ "9": "mwenda"
70
+ },
71
+ "initializer_range": 0.02,
72
+ "intermediate_size": 3072,
73
+ "label2id": {
74
+ "bbiri": 2,
75
+ "emu": 1,
76
+ "mukaaga": 6,
77
+ "munaana": 8,
78
+ "musanvu": 7,
79
+ "mwenda": 9,
80
+ "nnya": 4,
81
+ "notti": 0,
82
+ "ssatu": 3,
83
+ "ttaano": 5
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 sha256:83a905e0c6ced500f849e1c7a90338b1f01dfed272580dd770af2850ca20de56
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:eabad3d3e1d00694d28fdb452b4c4ad56eb8006527158f935c32dab90691986f
3
+ size 5841