hf-transformers-bot commited on
Commit
4606e56
·
verified ·
1 Parent(s): 43064b8

Update tiny models for MusicgenForConditionalGeneration

Browse files
config.json ADDED
@@ -0,0 +1,150 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "MusicgenForConditionalGeneration"
4
+ ],
5
+ "audio_encoder": {
6
+ "_name_or_path": "",
7
+ "architectures": null,
8
+ "audio_channels": 1,
9
+ "chunk_length_s": null,
10
+ "chunk_size_feed_forward": 0,
11
+ "codebook_dim": 99,
12
+ "codebook_size": 128,
13
+ "compress": 1,
14
+ "dilation_growth_rate": 2,
15
+ "dtype": null,
16
+ "hidden_size": 99,
17
+ "id2label": {
18
+ "0": "LABEL_0",
19
+ "1": "LABEL_1"
20
+ },
21
+ "is_encoder_decoder": false,
22
+ "kernel_size": 7,
23
+ "label2id": {
24
+ "LABEL_0": 0,
25
+ "LABEL_1": 1
26
+ },
27
+ "last_kernel_size": 7,
28
+ "model_type": "encodec",
29
+ "norm_type": "weight_norm",
30
+ "normalize": false,
31
+ "num_filters": 4,
32
+ "num_lstm_layers": 2,
33
+ "num_residual_layers": 1,
34
+ "output_attentions": false,
35
+ "output_hidden_states": false,
36
+ "overlap": null,
37
+ "pad_mode": "reflect",
38
+ "problem_type": null,
39
+ "residual_kernel_size": 3,
40
+ "return_dict": true,
41
+ "sampling_rate": 24000,
42
+ "target_bandwidths": [
43
+ 1.5,
44
+ 3.0,
45
+ 6.0,
46
+ 12.0,
47
+ 24.0
48
+ ],
49
+ "trim_right_ratio": 1.0,
50
+ "upsampling_ratios": [
51
+ 8,
52
+ 5,
53
+ 4,
54
+ 2
55
+ ],
56
+ "use_causal_conv": true,
57
+ "use_conv_shortcut": true
58
+ },
59
+ "decoder": {
60
+ "_name_or_path": "",
61
+ "activation_dropout": 0.0,
62
+ "activation_function": "gelu",
63
+ "add_cross_attention": false,
64
+ "architectures": null,
65
+ "attention_dropout": 0.0,
66
+ "audio_channels": 1,
67
+ "bos_token_id": 99,
68
+ "chunk_size_feed_forward": 0,
69
+ "cross_attention_hidden_size": null,
70
+ "decoder_start_token_id": 99,
71
+ "dropout": 0.1,
72
+ "dtype": null,
73
+ "eos_token_id": null,
74
+ "ffn_dim": 4,
75
+ "hidden_size": 16,
76
+ "id2label": {
77
+ "0": "LABEL_0",
78
+ "1": "LABEL_1"
79
+ },
80
+ "initializer_factor": 0.02,
81
+ "is_decoder": false,
82
+ "is_encoder_decoder": false,
83
+ "label2id": {
84
+ "LABEL_0": 0,
85
+ "LABEL_1": 1
86
+ },
87
+ "layerdrop": 0.0,
88
+ "max_position_embeddings": 2048,
89
+ "model_type": "musicgen_decoder",
90
+ "num_attention_heads": 4,
91
+ "num_codebooks": 4,
92
+ "num_hidden_layers": 2,
93
+ "output_attentions": false,
94
+ "output_hidden_states": false,
95
+ "pad_token_id": 99,
96
+ "problem_type": null,
97
+ "return_dict": true,
98
+ "scale_embedding": false,
99
+ "tie_word_embeddings": false,
100
+ "use_cache": true,
101
+ "vocab_size": 99
102
+ },
103
+ "dtype": "float32",
104
+ "initializer_factor": 0.02,
105
+ "is_encoder_decoder": true,
106
+ "model_type": "musicgen",
107
+ "text_encoder": {
108
+ "_name_or_path": "",
109
+ "architectures": null,
110
+ "chunk_size_feed_forward": 0,
111
+ "classifier_dropout": 0.0,
112
+ "d_ff": 4,
113
+ "d_kv": 64,
114
+ "d_model": 16,
115
+ "dense_act_fn": "relu",
116
+ "dropout_rate": 0.1,
117
+ "dtype": null,
118
+ "eos_token_id": 1,
119
+ "feed_forward_proj": "relu",
120
+ "id2label": {
121
+ "0": "LABEL_0",
122
+ "1": "LABEL_1"
123
+ },
124
+ "initializer_factor": 1.0,
125
+ "is_decoder": false,
126
+ "is_encoder_decoder": false,
127
+ "is_gated_act": false,
128
+ "label2id": {
129
+ "LABEL_0": 0,
130
+ "LABEL_1": 1
131
+ },
132
+ "layer_norm_epsilon": 1e-06,
133
+ "model_type": "t5",
134
+ "num_decoder_layers": 2,
135
+ "num_heads": 4,
136
+ "num_layers": 2,
137
+ "output_attentions": false,
138
+ "output_hidden_states": false,
139
+ "pad_token_id": 0,
140
+ "problem_type": null,
141
+ "relative_attention_max_distance": 128,
142
+ "relative_attention_num_buckets": 32,
143
+ "return_dict": true,
144
+ "scale_decoder_outputs": true,
145
+ "tie_word_embeddings": true,
146
+ "use_cache": false,
147
+ "vocab_size": 99
148
+ },
149
+ "transformers_version": "5.16.0.dev0"
150
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 99,
4
+ "decoder_start_token_id": 99,
5
+ "output_attentions": false,
6
+ "output_hidden_states": false,
7
+ "pad_token_id": 99,
8
+ "transformers_version": "5.16.0.dev0",
9
+ "use_cache": true
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a48de96bf54c42d5cc112b1acc17a4eb6da71c79c0d18b075ee2a0be1cf9acc8
3
+ size 6098268
preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length_s": null,
3
+ "feature_extractor_type": "EncodecFeatureExtractor",
4
+ "feature_size": 1,
5
+ "overlap": null,
6
+ "padding_side": "left",
7
+ "padding_value": 0.0,
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 32000
10
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "clean_up_tokenization_spaces": true,
4
+ "eos_token": "</s>",
5
+ "extra_ids": 100,
6
+ "extra_special_tokens": [
7
+ "<extra_id_0>",
8
+ "<extra_id_1>",
9
+ "<extra_id_2>",
10
+ "<extra_id_3>",
11
+ "<extra_id_4>",
12
+ "<extra_id_5>",
13
+ "<extra_id_6>",
14
+ "<extra_id_7>",
15
+ "<extra_id_8>",
16
+ "<extra_id_9>",
17
+ "<extra_id_10>",
18
+ "<extra_id_11>",
19
+ "<extra_id_12>",
20
+ "<extra_id_13>",
21
+ "<extra_id_14>",
22
+ "<extra_id_15>",
23
+ "<extra_id_16>",
24
+ "<extra_id_17>",
25
+ "<extra_id_18>",
26
+ "<extra_id_19>",
27
+ "<extra_id_20>",
28
+ "<extra_id_21>",
29
+ "<extra_id_22>",
30
+ "<extra_id_23>",
31
+ "<extra_id_24>",
32
+ "<extra_id_25>",
33
+ "<extra_id_26>",
34
+ "<extra_id_27>",
35
+ "<extra_id_28>",
36
+ "<extra_id_29>",
37
+ "<extra_id_30>",
38
+ "<extra_id_31>",
39
+ "<extra_id_32>",
40
+ "<extra_id_33>",
41
+ "<extra_id_34>",
42
+ "<extra_id_35>",
43
+ "<extra_id_36>",
44
+ "<extra_id_37>",
45
+ "<extra_id_38>",
46
+ "<extra_id_39>",
47
+ "<extra_id_40>",
48
+ "<extra_id_41>",
49
+ "<extra_id_42>",
50
+ "<extra_id_43>",
51
+ "<extra_id_44>",
52
+ "<extra_id_45>",
53
+ "<extra_id_46>",
54
+ "<extra_id_47>",
55
+ "<extra_id_48>",
56
+ "<extra_id_49>",
57
+ "<extra_id_50>",
58
+ "<extra_id_51>",
59
+ "<extra_id_52>",
60
+ "<extra_id_53>",
61
+ "<extra_id_54>",
62
+ "<extra_id_55>",
63
+ "<extra_id_56>",
64
+ "<extra_id_57>",
65
+ "<extra_id_58>",
66
+ "<extra_id_59>",
67
+ "<extra_id_60>",
68
+ "<extra_id_61>",
69
+ "<extra_id_62>",
70
+ "<extra_id_63>",
71
+ "<extra_id_64>",
72
+ "<extra_id_65>",
73
+ "<extra_id_66>",
74
+ "<extra_id_67>",
75
+ "<extra_id_68>",
76
+ "<extra_id_69>",
77
+ "<extra_id_70>",
78
+ "<extra_id_71>",
79
+ "<extra_id_72>",
80
+ "<extra_id_73>",
81
+ "<extra_id_74>",
82
+ "<extra_id_75>",
83
+ "<extra_id_76>",
84
+ "<extra_id_77>",
85
+ "<extra_id_78>",
86
+ "<extra_id_79>",
87
+ "<extra_id_80>",
88
+ "<extra_id_81>",
89
+ "<extra_id_82>",
90
+ "<extra_id_83>",
91
+ "<extra_id_84>",
92
+ "<extra_id_85>",
93
+ "<extra_id_86>",
94
+ "<extra_id_87>",
95
+ "<extra_id_88>",
96
+ "<extra_id_89>",
97
+ "<extra_id_90>",
98
+ "<extra_id_91>",
99
+ "<extra_id_92>",
100
+ "<extra_id_93>",
101
+ "<extra_id_94>",
102
+ "<extra_id_95>",
103
+ "<extra_id_96>",
104
+ "<extra_id_97>",
105
+ "<extra_id_98>",
106
+ "<extra_id_99>"
107
+ ],
108
+ "is_local": true,
109
+ "local_files_only": false,
110
+ "model_max_length": 512,
111
+ "pad_token": "<pad>",
112
+ "processor_class": "MusicgenProcessor",
113
+ "tokenizer_class": "T5Tokenizer",
114
+ "unk_token": "<unk>"
115
+ }