Upload JukeboxVQVAE
Browse files- config.json +227 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,227 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/jukebox-1b-lyrics",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"JukeboxVQVAE"
|
| 5 |
+
],
|
| 6 |
+
"cond_c_res": [
|
| 7 |
+
0,
|
| 8 |
+
1,
|
| 9 |
+
1
|
| 10 |
+
],
|
| 11 |
+
"cond_depth": [
|
| 12 |
+
3,
|
| 13 |
+
16,
|
| 14 |
+
16
|
| 15 |
+
],
|
| 16 |
+
"cond_dilation_cycle": [
|
| 17 |
+
null,
|
| 18 |
+
8,
|
| 19 |
+
8
|
| 20 |
+
],
|
| 21 |
+
"cond_dilation_growth_rate": [
|
| 22 |
+
1,
|
| 23 |
+
3,
|
| 24 |
+
3
|
| 25 |
+
],
|
| 26 |
+
"cond_downs_t": [
|
| 27 |
+
3,
|
| 28 |
+
2,
|
| 29 |
+
2
|
| 30 |
+
],
|
| 31 |
+
"cond_m_conv": 1,
|
| 32 |
+
"cond_res_scale": [
|
| 33 |
+
null,
|
| 34 |
+
true,
|
| 35 |
+
false
|
| 36 |
+
],
|
| 37 |
+
"cond_strides_t": [
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2
|
| 41 |
+
],
|
| 42 |
+
"cond_width": [
|
| 43 |
+
128,
|
| 44 |
+
1024,
|
| 45 |
+
1024
|
| 46 |
+
],
|
| 47 |
+
"cond_zero_out": false,
|
| 48 |
+
"copy_input": false,
|
| 49 |
+
"fp16_params": true,
|
| 50 |
+
"hop_fraction": [
|
| 51 |
+
0.125,
|
| 52 |
+
0.5,
|
| 53 |
+
0.5
|
| 54 |
+
],
|
| 55 |
+
"init_std": 0.2,
|
| 56 |
+
"lyric_conditioning": [
|
| 57 |
+
true,
|
| 58 |
+
false,
|
| 59 |
+
false
|
| 60 |
+
],
|
| 61 |
+
"lyric_enc_attn_dropout": 0.0,
|
| 62 |
+
"lyric_enc_attn_order": [
|
| 63 |
+
2,
|
| 64 |
+
0,
|
| 65 |
+
0
|
| 66 |
+
],
|
| 67 |
+
"lyric_enc_blocks": 32,
|
| 68 |
+
"lyric_enc_depth": [
|
| 69 |
+
18,
|
| 70 |
+
3,
|
| 71 |
+
3
|
| 72 |
+
],
|
| 73 |
+
"lyric_enc_emb_dropout": 0.0,
|
| 74 |
+
"lyric_enc_heads": 4,
|
| 75 |
+
"lyric_enc_init_scale": [
|
| 76 |
+
0.1,
|
| 77 |
+
0.4,
|
| 78 |
+
0.4
|
| 79 |
+
],
|
| 80 |
+
"lyric_enc_loss_fraction": [
|
| 81 |
+
0.4,
|
| 82 |
+
0.0,
|
| 83 |
+
0.0
|
| 84 |
+
],
|
| 85 |
+
"lyric_enc_m_attn": 0.25,
|
| 86 |
+
"lyric_enc_m_mlp": 1.0,
|
| 87 |
+
"lyric_enc_n_vocab": 79,
|
| 88 |
+
"lyric_enc_pos_init": false,
|
| 89 |
+
"lyric_enc_res_scale": false,
|
| 90 |
+
"lyric_enc_resid_dropout": 0.0,
|
| 91 |
+
"lyric_enc_spread": null,
|
| 92 |
+
"lyric_enc_width": [
|
| 93 |
+
128,
|
| 94 |
+
128,
|
| 95 |
+
128
|
| 96 |
+
],
|
| 97 |
+
"lyric_enc_zero_out": false,
|
| 98 |
+
"max_duration": 600.0,
|
| 99 |
+
"max_nb_genres": 1,
|
| 100 |
+
"merged_decoder": [
|
| 101 |
+
true,
|
| 102 |
+
false,
|
| 103 |
+
false
|
| 104 |
+
],
|
| 105 |
+
"metadata_conditioning": true,
|
| 106 |
+
"metadata_dims": [
|
| 107 |
+
[
|
| 108 |
+
604,
|
| 109 |
+
7898
|
| 110 |
+
],
|
| 111 |
+
[
|
| 112 |
+
120,
|
| 113 |
+
4111
|
| 114 |
+
],
|
| 115 |
+
[
|
| 116 |
+
120,
|
| 117 |
+
4111
|
| 118 |
+
]
|
| 119 |
+
],
|
| 120 |
+
"min_duration": 0,
|
| 121 |
+
"model_type": "jukebox",
|
| 122 |
+
"nb_priors": 3,
|
| 123 |
+
"nb_relevant_lyric_tokens": [
|
| 124 |
+
384,
|
| 125 |
+
0,
|
| 126 |
+
0
|
| 127 |
+
],
|
| 128 |
+
"prior_alignment_head": [
|
| 129 |
+
2,
|
| 130 |
+
null,
|
| 131 |
+
null
|
| 132 |
+
],
|
| 133 |
+
"prior_alignment_layer": [
|
| 134 |
+
68,
|
| 135 |
+
null,
|
| 136 |
+
null
|
| 137 |
+
],
|
| 138 |
+
"prior_attn_dropout": 0,
|
| 139 |
+
"prior_attn_order": [
|
| 140 |
+
12,
|
| 141 |
+
2,
|
| 142 |
+
2
|
| 143 |
+
],
|
| 144 |
+
"prior_blocks": 64,
|
| 145 |
+
"prior_depth": [
|
| 146 |
+
72,
|
| 147 |
+
72,
|
| 148 |
+
72
|
| 149 |
+
],
|
| 150 |
+
"prior_emb_dropout": 0,
|
| 151 |
+
"prior_init_scale": [
|
| 152 |
+
0.2,
|
| 153 |
+
1,
|
| 154 |
+
1
|
| 155 |
+
],
|
| 156 |
+
"prior_latent_dim": 2048,
|
| 157 |
+
"prior_m_attn": 0.25,
|
| 158 |
+
"prior_n_ctx": [
|
| 159 |
+
6144,
|
| 160 |
+
8192,
|
| 161 |
+
8192
|
| 162 |
+
],
|
| 163 |
+
"prior_n_heads": [
|
| 164 |
+
2,
|
| 165 |
+
1,
|
| 166 |
+
1
|
| 167 |
+
],
|
| 168 |
+
"prior_pos_init": false,
|
| 169 |
+
"prior_res_scale": false,
|
| 170 |
+
"prior_resid_dropout": 0,
|
| 171 |
+
"prior_spread": null,
|
| 172 |
+
"prior_width": [
|
| 173 |
+
2048,
|
| 174 |
+
1920,
|
| 175 |
+
1920
|
| 176 |
+
],
|
| 177 |
+
"prior_zero_out": false,
|
| 178 |
+
"sample_length": 1058304,
|
| 179 |
+
"sampling_rate": 44100,
|
| 180 |
+
"single_enc_dec": [
|
| 181 |
+
true,
|
| 182 |
+
false,
|
| 183 |
+
false
|
| 184 |
+
],
|
| 185 |
+
"timing_dims": 64,
|
| 186 |
+
"torch_dtype": "float32",
|
| 187 |
+
"transformers_version": "4.24.0.dev0",
|
| 188 |
+
"vqvae_codebook_dimension": 2048,
|
| 189 |
+
"vqvae_commit": 0.02,
|
| 190 |
+
"vqvae_conv_block_depth": 4,
|
| 191 |
+
"vqvae_conv_block_width": 32,
|
| 192 |
+
"vqvae_depth": 4,
|
| 193 |
+
"vqvae_dilation_cycle": null,
|
| 194 |
+
"vqvae_dilation_growth_rate": 3,
|
| 195 |
+
"vqvae_downs_t": [
|
| 196 |
+
3,
|
| 197 |
+
2,
|
| 198 |
+
2
|
| 199 |
+
],
|
| 200 |
+
"vqvae_emmbedding_width": 64,
|
| 201 |
+
"vqvae_levels": 3,
|
| 202 |
+
"vqvae_lmu": 0.99,
|
| 203 |
+
"vqvae_m_conv": 1,
|
| 204 |
+
"vqvae_multipliers": [
|
| 205 |
+
2,
|
| 206 |
+
1,
|
| 207 |
+
1
|
| 208 |
+
],
|
| 209 |
+
"vqvae_music_tokens_shapes": [
|
| 210 |
+
[
|
| 211 |
+
8268
|
| 212 |
+
],
|
| 213 |
+
[
|
| 214 |
+
33072
|
| 215 |
+
],
|
| 216 |
+
[
|
| 217 |
+
132288
|
| 218 |
+
]
|
| 219 |
+
],
|
| 220 |
+
"vqvae_reverse_decoder_dilation": 1,
|
| 221 |
+
"vqvae_strides_t": [
|
| 222 |
+
2,
|
| 223 |
+
2,
|
| 224 |
+
2
|
| 225 |
+
],
|
| 226 |
+
"vqvae_width": 32
|
| 227 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0fe9920ae68298eb3fd3598d53062fc004f73aa361ee7be01b0d7eb3dec7232
|
| 3 |
+
size 7745991
|