Upload folder using huggingface_hub
Browse files- config.json +48 -0
- model.safetensors +3 -0
- preprocessor_config.json +52 -0
config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma3nAudioEncoder"
|
| 4 |
+
],
|
| 5 |
+
"conf_attention_chunk_size": 12,
|
| 6 |
+
"conf_attention_context_left": 13,
|
| 7 |
+
"conf_attention_context_right": 0,
|
| 8 |
+
"conf_attention_logit_cap": 50.0,
|
| 9 |
+
"conf_conv_kernel_size": 5,
|
| 10 |
+
"conf_num_attention_heads": 8,
|
| 11 |
+
"conf_num_hidden_layers": 12,
|
| 12 |
+
"conf_reduction_factor": 4,
|
| 13 |
+
"conf_residual_weight": 0.5,
|
| 14 |
+
"gradient_clipping": 10000000000.0,
|
| 15 |
+
"hidden_size": 1536,
|
| 16 |
+
"input_feat_size": 128,
|
| 17 |
+
"model_type": "gemma3n_audio",
|
| 18 |
+
"rms_norm_eps": 1e-06,
|
| 19 |
+
"sscp_conv_channel_size": [
|
| 20 |
+
128,
|
| 21 |
+
32
|
| 22 |
+
],
|
| 23 |
+
"sscp_conv_group_norm_eps": 0.001,
|
| 24 |
+
"sscp_conv_kernel_size": [
|
| 25 |
+
[
|
| 26 |
+
3,
|
| 27 |
+
3
|
| 28 |
+
],
|
| 29 |
+
[
|
| 30 |
+
3,
|
| 31 |
+
3
|
| 32 |
+
]
|
| 33 |
+
],
|
| 34 |
+
"sscp_conv_stride_size": [
|
| 35 |
+
[
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
[
|
| 40 |
+
2,
|
| 41 |
+
2
|
| 42 |
+
]
|
| 43 |
+
],
|
| 44 |
+
"torch_dtype": "float32",
|
| 45 |
+
"transformers_version": "4.53.2",
|
| 46 |
+
"vocab_offset": 262272,
|
| 47 |
+
"vocab_size": 128
|
| 48 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c07b48dc2c7e5bd5a41f74551c337dcfb2e1eaf34d78ad2c52739e05832a4942
|
| 3 |
+
size 2725426224
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": null,
|
| 3 |
+
"data_format": "channels_first",
|
| 4 |
+
"default_to_square": false,
|
| 5 |
+
"device": null,
|
| 6 |
+
"disable_grouping": null,
|
| 7 |
+
"dither": 0.0,
|
| 8 |
+
"do_center_crop": null,
|
| 9 |
+
"do_convert_rgb": null,
|
| 10 |
+
"do_normalize": false,
|
| 11 |
+
"do_rescale": true,
|
| 12 |
+
"do_resize": true,
|
| 13 |
+
"feature_extractor_type": "Gemma3nAudioFeatureExtractor",
|
| 14 |
+
"feature_size": 128,
|
| 15 |
+
"fft_length": 1024,
|
| 16 |
+
"fft_overdrive": true,
|
| 17 |
+
"frame_length": 512,
|
| 18 |
+
"hop_length": 160,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.5,
|
| 21 |
+
0.5,
|
| 22 |
+
0.5
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "SiglipImageProcessorFast",
|
| 25 |
+
"image_seq_length": 256,
|
| 26 |
+
"image_std": [
|
| 27 |
+
0.5,
|
| 28 |
+
0.5,
|
| 29 |
+
0.5
|
| 30 |
+
],
|
| 31 |
+
"input_data_format": null,
|
| 32 |
+
"input_scale_factor": 1.0,
|
| 33 |
+
"max_frequency": 7600.0,
|
| 34 |
+
"mel_floor": 1e-05,
|
| 35 |
+
"min_frequency": 125.0,
|
| 36 |
+
"padding_side": "right",
|
| 37 |
+
"padding_value": 0.0,
|
| 38 |
+
"per_bin_mean": null,
|
| 39 |
+
"per_bin_stddev": null,
|
| 40 |
+
"preemphasis": 0.97,
|
| 41 |
+
"preemphasis_htk_flavor": true,
|
| 42 |
+
"processor_class": "Gemma3nProcessor",
|
| 43 |
+
"resample": 2,
|
| 44 |
+
"rescale_factor": 0.00392156862745098,
|
| 45 |
+
"return_attention_mask": true,
|
| 46 |
+
"return_tensors": null,
|
| 47 |
+
"sampling_rate": 16000,
|
| 48 |
+
"size": {
|
| 49 |
+
"height": 768,
|
| 50 |
+
"width": 768
|
| 51 |
+
}
|
| 52 |
+
}
|