| { | |
| "architectures": [ | |
| "ASTModel" | |
| ], | |
| "attention_probs_dropout_prob": 0.0, | |
| "frequency_stride": 10, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.0, | |
| "hidden_size": 32, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 37, | |
| "layer_norm_eps": 1e-12, | |
| "max_length": 1024, | |
| "model_type": "audio-spectrogram-transformer", | |
| "num_attention_heads": 4, | |
| "num_hidden_layers": 5, | |
| "num_mel_bins": 128, | |
| "patch_size": 1, | |
| "qkv_bias": true, | |
| "time_stride": 10, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.25.1" | |
| } | |