Upload GroundingDinoForObjectDetection
Browse files- config.json +82 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./grounding_dino_screen_ai_model",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "relu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"GroundingDinoForObjectDetection"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"auxiliary_loss": false,
|
| 10 |
+
"backbone": null,
|
| 11 |
+
"backbone_config": {
|
| 12 |
+
"depths": [
|
| 13 |
+
2,
|
| 14 |
+
2,
|
| 15 |
+
18,
|
| 16 |
+
2
|
| 17 |
+
],
|
| 18 |
+
"embed_dim": 128,
|
| 19 |
+
"hidden_size": 1024,
|
| 20 |
+
"image_size": 384,
|
| 21 |
+
"model_type": "swin",
|
| 22 |
+
"num_heads": [
|
| 23 |
+
4,
|
| 24 |
+
8,
|
| 25 |
+
16,
|
| 26 |
+
32
|
| 27 |
+
],
|
| 28 |
+
"out_features": [
|
| 29 |
+
"stage2",
|
| 30 |
+
"stage3",
|
| 31 |
+
"stage4"
|
| 32 |
+
],
|
| 33 |
+
"out_indices": [
|
| 34 |
+
2,
|
| 35 |
+
3,
|
| 36 |
+
4
|
| 37 |
+
],
|
| 38 |
+
"window_size": 12
|
| 39 |
+
},
|
| 40 |
+
"backbone_kwargs": null,
|
| 41 |
+
"bbox_cost": 5.0,
|
| 42 |
+
"bbox_loss_coefficient": 5.0,
|
| 43 |
+
"class_cost": 1.0,
|
| 44 |
+
"d_model": 256,
|
| 45 |
+
"decoder_attention_heads": 8,
|
| 46 |
+
"decoder_bbox_embed_share": true,
|
| 47 |
+
"decoder_ffn_dim": 2048,
|
| 48 |
+
"decoder_layers": 6,
|
| 49 |
+
"decoder_n_points": 4,
|
| 50 |
+
"disable_custom_kernels": false,
|
| 51 |
+
"dropout": 0.1,
|
| 52 |
+
"embedding_init_target": true,
|
| 53 |
+
"encoder_attention_heads": 8,
|
| 54 |
+
"encoder_ffn_dim": 2048,
|
| 55 |
+
"encoder_layers": 6,
|
| 56 |
+
"encoder_n_points": 4,
|
| 57 |
+
"focal_alpha": 0.25,
|
| 58 |
+
"fusion_dropout": 0.0,
|
| 59 |
+
"fusion_droppath": 0.1,
|
| 60 |
+
"giou_cost": 2.0,
|
| 61 |
+
"giou_loss_coefficient": 2.0,
|
| 62 |
+
"init_std": 0.02,
|
| 63 |
+
"is_encoder_decoder": true,
|
| 64 |
+
"layer_norm_eps": 1e-05,
|
| 65 |
+
"max_text_len": 256,
|
| 66 |
+
"model_type": "grounding-dino",
|
| 67 |
+
"num_feature_levels": 4,
|
| 68 |
+
"num_queries": 900,
|
| 69 |
+
"position_embedding_type": "sine",
|
| 70 |
+
"positional_embedding_temperature": 20,
|
| 71 |
+
"query_dim": 4,
|
| 72 |
+
"text_config": {
|
| 73 |
+
"model_type": "bert"
|
| 74 |
+
},
|
| 75 |
+
"text_enhancer_dropout": 0.0,
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.41.2",
|
| 78 |
+
"two_stage": true,
|
| 79 |
+
"two_stage_bbox_embed_share": false,
|
| 80 |
+
"use_pretrained_backbone": false,
|
| 81 |
+
"use_timm_backbone": false
|
| 82 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fad033ad24f64554e9ebc05f563875f3df732a4f090b6ced80dbf1ec95882870
|
| 3 |
+
size 933400776
|