config.json
| 1 | { |
| 2 | "architectures": [ |
| 3 | "ViTForImageClassification" |
| 4 | ], |
| 5 | "attention_probs_dropout_prob": 0.0, |
| 6 | "encoder_stride": 16, |
| 7 | "hidden_act": "gelu", |
| 8 | "hidden_dropout_prob": 0.0, |
| 9 | "hidden_size": 384, |
| 10 | "image_size": 384, |
| 11 | "initializer_range": 0.02, |
| 12 | "intermediate_size": 3072, |
| 13 | "layer_norm_eps": 1e-06, |
| 14 | "mlp_ratio": 4, |
| 15 | "model_type": "vit", |
| 16 | "num_attention_heads": 12, |
| 17 | "num_channels": 3, |
| 18 | "num_classes": 1, |
| 19 | "num_heads": 6, |
| 20 | "num_hidden_layers": 12, |
| 21 | "num_layers": 12, |
| 22 | "patch_size": 16, |
| 23 | "qkv_bias": true, |
| 24 | "torch_dtype": "float32", |
| 25 | "transformers_version": "4.50.0.dev0" |
| 26 | } |
| 27 | |