nrshoudi commited on
Commit
ced73b3
1 Parent(s): 106fd65

End of training

Browse files
README.md ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: facebook/hubert-large-ll60k
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - wer
8
+ model-index:
9
+ - name: HuBERT_Jibbali_lang
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # HuBERT_Jibbali_lang
17
+
18
+ This model is a fine-tuned version of [facebook/hubert-large-ll60k](https://huggingface.co/facebook/hubert-large-ll60k) on an unknown dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.2017
21
+ - Wer: 0.1944
22
+ - Cet: 0.1189
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: 0.0003
42
+ - train_batch_size: 16
43
+ - eval_batch_size: 8
44
+ - seed: 42
45
+ - gradient_accumulation_steps: 2
46
+ - total_train_batch_size: 32
47
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
+ - lr_scheduler_type: linear
49
+ - lr_scheduler_warmup_steps: 500
50
+ - num_epochs: 20
51
+ - mixed_precision_training: Native AMP
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cet |
56
+ |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
57
+ | 10.6563 | 0.99 | 56 | 5.6577 | 1.0 | 0.9812 |
58
+ | 3.3895 | 2.0 | 113 | 3.2018 | 1.0 | 0.9812 |
59
+ | 3.1588 | 2.99 | 169 | 3.1347 | 1.0 | 0.9812 |
60
+ | 3.1308 | 4.0 | 226 | 3.0567 | 1.0 | 0.9812 |
61
+ | 2.8933 | 4.99 | 282 | 2.8226 | 1.0 | 0.9353 |
62
+ | 2.5444 | 6.0 | 339 | 2.0947 | 1.0 | 0.8588 |
63
+ | 0.995 | 6.99 | 395 | 0.5049 | 0.4974 | 0.1654 |
64
+ | 0.3567 | 8.0 | 452 | 0.2622 | 0.2485 | 0.1132 |
65
+ | 0.2914 | 8.99 | 508 | 0.1980 | 0.2105 | 0.0749 |
66
+ | 0.14 | 10.0 | 565 | 0.2154 | 0.2069 | 0.0821 |
67
+ | 0.1442 | 10.99 | 621 | 0.1965 | 0.1988 | 0.0969 |
68
+ | 0.1401 | 12.0 | 678 | 0.2135 | 0.1937 | 0.0960 |
69
+ | 0.1019 | 12.99 | 734 | 0.2185 | 0.1948 | 0.1094 |
70
+ | 0.1088 | 14.0 | 791 | 0.1957 | 0.1966 | 0.1121 |
71
+ | 0.1314 | 14.99 | 847 | 0.1983 | 0.1933 | 0.1019 |
72
+ | 0.0522 | 16.0 | 904 | 0.2026 | 0.1944 | 0.1258 |
73
+ | 0.126 | 16.99 | 960 | 0.2033 | 0.1944 | 0.1142 |
74
+ | 0.1028 | 18.0 | 1017 | 0.1940 | 0.1974 | 0.1158 |
75
+ | 0.0767 | 18.99 | 1073 | 0.1969 | 0.1948 | 0.1149 |
76
+ | 0.0468 | 19.82 | 1120 | 0.2017 | 0.1944 | 0.1189 |
77
+
78
+
79
+ ### Framework versions
80
+
81
+ - Transformers 4.35.2
82
+ - Pytorch 2.1.0+cu121
83
+ - Datasets 2.17.0
84
+ - Tokenizers 0.15.2
config.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/hubert-large-ll60k",
3
+ "activation_dropout": 0.0,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "HubertForCTC"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": true,
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": "mean",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": true,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_dropout": 0.0,
45
+ "feat_extract_norm": "layer",
46
+ "feat_proj_dropout": 0.0,
47
+ "feat_proj_layer_norm": true,
48
+ "final_dropout": 0.0,
49
+ "gradient_checkpointing": false,
50
+ "hidden_act": "gelu",
51
+ "hidden_dropout": 0.1,
52
+ "hidden_size": 1024,
53
+ "initializer_range": 0.02,
54
+ "intermediate_size": 4096,
55
+ "layer_norm_eps": 1e-05,
56
+ "layerdrop": 0.1,
57
+ "mask_channel_length": 10,
58
+ "mask_channel_min_space": 1,
59
+ "mask_channel_other": 0.0,
60
+ "mask_channel_prob": 0.0,
61
+ "mask_channel_selection": "static",
62
+ "mask_feature_length": 10,
63
+ "mask_feature_min_masks": 0,
64
+ "mask_feature_prob": 0.0,
65
+ "mask_time_length": 10,
66
+ "mask_time_min_masks": 2,
67
+ "mask_time_min_space": 1,
68
+ "mask_time_other": 0.0,
69
+ "mask_time_prob": 0.05,
70
+ "mask_time_selection": "static",
71
+ "model_type": "hubert",
72
+ "num_attention_heads": 16,
73
+ "num_conv_pos_embedding_groups": 16,
74
+ "num_conv_pos_embeddings": 128,
75
+ "num_feat_extract_layers": 7,
76
+ "num_hidden_layers": 24,
77
+ "pad_token_id": 37,
78
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
79
+ "torch_dtype": "float32",
80
+ "transformers_version": "4.35.2",
81
+ "use_weighted_layer_sum": false,
82
+ "vocab_size": 40
83
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd1cda469afbc419b57dfd33335f194d56b91e0621b754b8e4e0a553ad35e4e4
3
+ size 1261970648
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": true,
8
+ "sampling_rate": 16000
9
+ }
runs/Feb17_14-38-37_96309f846a0a/events.out.tfevents.1708180755.96309f846a0a.1921.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d06d10098a2c49a3099f26861ed26ec265cb860f76679a2f858c8de384d3fe0
3
+ size 30541
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6304f3c17299b352b176eac943fb7ae7e6dc2637ad5228c256662027a27a92ec
3
+ size 4600