Spaces:
Sleeping
Sleeping
File size: 10,951 Bytes
7417a6a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 | Model:
TextConditionedSeparator(
(clap): ClapModel(
(text_model): ClapTextModel(
(embeddings): ClapTextEmbeddings(
(word_embeddings): Embedding(50265, 768, padding_idx=1)
(position_embeddings): Embedding(514, 768, padding_idx=1)
(token_type_embeddings): Embedding(1, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): ClapTextEncoder(
(layer): ModuleList(
(0-11): 12 x ClapTextLayer(
(attention): ClapTextAttention(
(self): ClapTextSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): ClapTextSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): ClapTextIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ClapTextOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): ClapTextPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
(text_projection): ClapProjectionLayer(
(linear1): Linear(in_features=768, out_features=512, bias=True)
(activation): ReLU()
(linear2): Linear(in_features=512, out_features=512, bias=True)
)
(audio_model): ClapAudioModel(
(audio_encoder): ClapAudioEncoder(
(patch_embed): ClapAudioPatchEmbed(
(proj): Conv2d(1, 96, kernel_size=(4, 4), stride=(4, 4))
(norm): LayerNorm((96,), eps=1e-05, elementwise_affine=True)
)
(layers): ModuleList(
(0): ClapAudioStage(
(blocks): ModuleList(
(0-1): 2 x ClapAudioLayer(
(layernorm_before): LayerNorm((96,), eps=1e-05, elementwise_affine=True)
(attention): ClapAudioAttention(
(self): ClapAudioSelfAttention(
(query): Linear(in_features=96, out_features=96, bias=True)
(key): Linear(in_features=96, out_features=96, bias=True)
(value): Linear(in_features=96, out_features=96, bias=True)
(dropout): Dropout(p=0.0, inplace=False)
)
(output): ClapAudioSelfOutput(
(dense): Linear(in_features=96, out_features=96, bias=True)
(dropout): Dropout(p=0.0, inplace=False)
)
)
(drop_path): Identity()
(layernorm_after): LayerNorm((96,), eps=1e-05, elementwise_affine=True)
(intermediate): ClapAudioIntermediate(
(dense): Linear(in_features=96, out_features=384, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ClapAudioOutput(
(dense): Linear(in_features=384, out_features=96, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(downsample): ClapAudioPatchMerging(
(reduction): Linear(in_features=384, out_features=192, bias=False)
(norm): LayerNorm((384,), eps=1e-05, elementwise_affine=True)
)
)
(1): ClapAudioStage(
(blocks): ModuleList(
(0-1): 2 x ClapAudioLayer(
(layernorm_before): LayerNorm((192,), eps=1e-05, elementwise_affine=True)
(attention): ClapAudioAttention(
(self): ClapAudioSelfAttention(
(query): Linear(in_features=192, out_features=192, bias=True)
(key): Linear(in_features=192, out_features=192, bias=True)
(value): Linear(in_features=192, out_features=192, bias=True)
(dropout): Dropout(p=0.0, inplace=False)
)
(output): ClapAudioSelfOutput(
(dense): Linear(in_features=192, out_features=192, bias=True)
(dropout): Dropout(p=0.0, inplace=False)
)
)
(drop_path): Identity()
(layernorm_after): LayerNorm((192,), eps=1e-05, elementwise_affine=True)
(intermediate): ClapAudioIntermediate(
(dense): Linear(in_features=192, out_features=768, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ClapAudioOutput(
(dense): Linear(in_features=768, out_features=192, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(downsample): ClapAudioPatchMerging(
(reduction): Linear(in_features=768, out_features=384, bias=False)
(norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
)
)
(2): ClapAudioStage(
(blocks): ModuleList(
(0-5): 6 x ClapAudioLayer(
(layernorm_before): LayerNorm((384,), eps=1e-05, elementwise_affine=True)
(attention): ClapAudioAttention(
(self): ClapAudioSelfAttention(
(query): Linear(in_features=384, out_features=384, bias=True)
(key): Linear(in_features=384, out_features=384, bias=True)
(value): Linear(in_features=384, out_features=384, bias=True)
(dropout): Dropout(p=0.0, inplace=False)
)
(output): ClapAudioSelfOutput(
(dense): Linear(in_features=384, out_features=384, bias=True)
(dropout): Dropout(p=0.0, inplace=False)
)
)
(drop_path): Identity()
(layernorm_after): LayerNorm((384,), eps=1e-05, elementwise_affine=True)
(intermediate): ClapAudioIntermediate(
(dense): Linear(in_features=384, out_features=1536, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ClapAudioOutput(
(dense): Linear(in_features=1536, out_features=384, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(downsample): ClapAudioPatchMerging(
(reduction): Linear(in_features=1536, out_features=768, bias=False)
(norm): LayerNorm((1536,), eps=1e-05, elementwise_affine=True)
)
)
(3): ClapAudioStage(
(blocks): ModuleList(
(0-1): 2 x ClapAudioLayer(
(layernorm_before): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attention): ClapAudioAttention(
(self): ClapAudioSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.0, inplace=False)
)
(output): ClapAudioSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.0, inplace=False)
)
)
(drop_path): Identity()
(layernorm_after): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(intermediate): ClapAudioIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ClapAudioOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
)
(batch_norm): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(avgpool): AdaptiveAvgPool1d(output_size=1)
)
)
(audio_projection): ClapProjectionLayer(
(linear1): Linear(in_features=768, out_features=512, bias=True)
(activation): ReLU()
(linear2): Linear(in_features=512, out_features=512, bias=True)
)
)
(z_encoder): PatchConv1d(
(conv): Conv1d(1, 256, kernel_size=(16,), stride=(8,))
)
(text_proj): Linear(in_features=512, out_features=256, bias=True)
(z_proj): Linear(in_features=256, out_features=256, bias=True)
(cross): CrossAttention(
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=256, out_features=256, bias=True)
)
(ln1): LayerNorm((256,), eps=1e-05, elementwise_affine=True)
(ff): MLP(
(fc1): Linear(in_features=256, out_features=1024, bias=True)
(fc2): Linear(in_features=1024, out_features=256, bias=True)
(act): GELU(approximate='none')
)
(ln2): LayerNorm((256,), eps=1e-05, elementwise_affine=True)
)
(transformer): TransformerEncoder(
(layers): ModuleList(
(0-5): 6 x TransformerEncoderLayer(
(self_attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=256, out_features=256, bias=True)
)
(linear1): Linear(in_features=256, out_features=1024, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
(linear2): Linear(in_features=1024, out_features=256, bias=True)
(norm1): LayerNorm((256,), eps=1e-05, elementwise_affine=True)
(norm2): LayerNorm((256,), eps=1e-05, elementwise_affine=True)
(dropout1): Dropout(p=0.1, inplace=False)
(dropout2): Dropout(p=0.1, inplace=False)
)
)
)
(spec_decoder): Sequential(
(0): Linear(in_features=256, out_features=256, bias=True)
(1): GELU(approximate='none')
(2): Linear(in_features=256, out_features=2049, bias=True)
)
)
output waveform shape: torch.Size([2, 1, 48000])
output spectrogram shape: torch.Size([2, 12001, 2049])
|