aryamtos commited on
Commit
1aff212
1 Parent(s): 7fd07db

End of training

Browse files
Files changed (5) hide show
  1. README.md +61 -0
  2. config.json +191 -0
  3. model.safetensors +0 -0
  4. preprocessor_config.json +10 -0
  5. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: facebook/wav2vec2-conformer-rel-pos-large
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - speech_commands
8
+ model-index:
9
+ - name: wav2vec2-conformer-rel-pos-large-speech-commands
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
+ # wav2vec2-conformer-rel-pos-large-speech-commands
17
+
18
+ This model is a fine-tuned version of [facebook/wav2vec2-conformer-rel-pos-large](https://huggingface.co/facebook/wav2vec2-conformer-rel-pos-large) on the speech_commands dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - eval_loss: 0.5969
21
+ - eval_accuracy: 0.9678
22
+ - eval_runtime: 128.1995
23
+ - eval_samples_per_second: 77.863
24
+ - eval_steps_per_second: 1.217
25
+ - epoch: 9.98
26
+ - step: 3310
27
+
28
+ ## Model description
29
+
30
+ More information needed
31
+
32
+ ## Intended uses & limitations
33
+
34
+ More information needed
35
+
36
+ ## Training and evaluation data
37
+
38
+ More information needed
39
+
40
+ ## Training procedure
41
+
42
+ ### Training hyperparameters
43
+
44
+ The following hyperparameters were used during training:
45
+ - learning_rate: 1e-05
46
+ - train_batch_size: 64
47
+ - eval_batch_size: 64
48
+ - seed: 42
49
+ - gradient_accumulation_steps: 4
50
+ - total_train_batch_size: 256
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: linear
53
+ - lr_scheduler_warmup_ratio: 0.1
54
+ - num_epochs: 10
55
+
56
+ ### Framework versions
57
+
58
+ - Transformers 4.35.0
59
+ - Pytorch 2.1.0+cu121
60
+ - Datasets 2.14.7.dev0
61
+ - Tokenizers 0.14.1
config.json ADDED
@@ -0,0 +1,191 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-conformer-rel-pos-large",
3
+ "activation_dropout": 0.1,
4
+ "adapter_kernel_size": 3,
5
+ "adapter_stride": 2,
6
+ "add_adapter": false,
7
+ "apply_spec_augment": true,
8
+ "architectures": [
9
+ "Wav2Vec2ConformerForSequenceClassification"
10
+ ],
11
+ "attention_dropout": 0.1,
12
+ "bos_token_id": 1,
13
+ "classifier_proj_size": 256,
14
+ "codevector_dim": 768,
15
+ "conformer_conv_dropout": 0.1,
16
+ "contrastive_logits_temperature": 0.1,
17
+ "conv_bias": true,
18
+ "conv_depthwise_kernel_size": 31,
19
+ "conv_dim": [
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512,
25
+ 512,
26
+ 512
27
+ ],
28
+ "conv_kernel": [
29
+ 10,
30
+ 3,
31
+ 3,
32
+ 3,
33
+ 3,
34
+ 2,
35
+ 2
36
+ ],
37
+ "conv_stride": [
38
+ 5,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2,
43
+ 2,
44
+ 2
45
+ ],
46
+ "ctc_loss_reduction": "sum",
47
+ "ctc_zero_infinity": false,
48
+ "diversity_loss_weight": 0.1,
49
+ "do_stable_layer_norm": true,
50
+ "eos_token_id": 2,
51
+ "feat_extract_activation": "gelu",
52
+ "feat_extract_dropout": 0.0,
53
+ "feat_extract_norm": "layer",
54
+ "feat_proj_dropout": 0.1,
55
+ "feat_quantizer_dropout": 0.0,
56
+ "final_dropout": 0.1,
57
+ "gradient_checkpointing": false,
58
+ "hidden_act": "swish",
59
+ "hidden_dropout": 0.1,
60
+ "hidden_dropout_prob": 0.1,
61
+ "hidden_size": 1024,
62
+ "id2label": {
63
+ "0": "yes",
64
+ "1": "no",
65
+ "10": "zero",
66
+ "11": "one",
67
+ "12": "two",
68
+ "13": "three",
69
+ "14": "four",
70
+ "15": "five",
71
+ "16": "six",
72
+ "17": "seven",
73
+ "18": "eight",
74
+ "19": "nine",
75
+ "2": "up",
76
+ "20": "bed",
77
+ "21": "bird",
78
+ "22": "cat",
79
+ "23": "dog",
80
+ "24": "happy",
81
+ "25": "house",
82
+ "26": "marvin",
83
+ "27": "sheila",
84
+ "28": "tree",
85
+ "29": "wow",
86
+ "3": "down",
87
+ "30": "backward",
88
+ "31": "forward",
89
+ "32": "follow",
90
+ "33": "learn",
91
+ "34": "visual",
92
+ "35": "_silence_",
93
+ "4": "left",
94
+ "5": "right",
95
+ "6": "on",
96
+ "7": "off",
97
+ "8": "stop",
98
+ "9": "go"
99
+ },
100
+ "initializer_range": 0.02,
101
+ "intermediate_size": 4096,
102
+ "label2id": {
103
+ "_silence_": 35,
104
+ "backward": 30,
105
+ "bed": 20,
106
+ "bird": 21,
107
+ "cat": 22,
108
+ "dog": 23,
109
+ "down": 3,
110
+ "eight": 18,
111
+ "five": 15,
112
+ "follow": 32,
113
+ "forward": 31,
114
+ "four": 14,
115
+ "go": 9,
116
+ "happy": 24,
117
+ "house": 25,
118
+ "learn": 33,
119
+ "left": 4,
120
+ "marvin": 26,
121
+ "nine": 19,
122
+ "no": 1,
123
+ "off": 7,
124
+ "on": 6,
125
+ "one": 11,
126
+ "right": 5,
127
+ "seven": 17,
128
+ "sheila": 27,
129
+ "six": 16,
130
+ "stop": 8,
131
+ "three": 13,
132
+ "tree": 28,
133
+ "two": 12,
134
+ "up": 2,
135
+ "visual": 34,
136
+ "wow": 29,
137
+ "yes": 0,
138
+ "zero": 10
139
+ },
140
+ "layer_norm_eps": 1e-05,
141
+ "layerdrop": 0.0,
142
+ "mask_feature_length": 10,
143
+ "mask_feature_min_masks": 0,
144
+ "mask_feature_prob": 0.0,
145
+ "mask_time_length": 10,
146
+ "mask_time_min_masks": 2,
147
+ "mask_time_prob": 0.05,
148
+ "max_source_positions": 5000,
149
+ "model_type": "wav2vec2-conformer",
150
+ "num_adapter_layers": 3,
151
+ "num_attention_heads": 16,
152
+ "num_codevector_groups": 2,
153
+ "num_codevectors_per_group": 320,
154
+ "num_conv_pos_embedding_groups": 16,
155
+ "num_conv_pos_embeddings": 128,
156
+ "num_feat_extract_layers": 7,
157
+ "num_hidden_layers": 24,
158
+ "num_negatives": 100,
159
+ "output_hidden_size": 1024,
160
+ "pad_token_id": 0,
161
+ "position_embeddings_type": "relative",
162
+ "proj_codevector_dim": 768,
163
+ "rotary_embedding_base": 10000,
164
+ "tdnn_dilation": [
165
+ 1,
166
+ 2,
167
+ 3,
168
+ 1,
169
+ 1
170
+ ],
171
+ "tdnn_dim": [
172
+ 512,
173
+ 512,
174
+ 512,
175
+ 512,
176
+ 1500
177
+ ],
178
+ "tdnn_kernel": [
179
+ 5,
180
+ 3,
181
+ 3,
182
+ 1,
183
+ 1
184
+ ],
185
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
186
+ "torch_dtype": "float32",
187
+ "transformers_version": "4.35.0",
188
+ "use_weighted_layer_sum": false,
189
+ "vocab_size": null,
190
+ "xvector_output_dim": 512
191
+ }
model.safetensors ADDED
File without changes
preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37b18d09882bc4e56732f33f2f244e57ec4f7c8cd27ea61031d33aa8e70f63ac
3
+ size 4728