Hugging Face's logo

ConservationDrones
/
DroneMegaDetector

Object Detection
Transformers
Safetensors
PyTorch
rf_detr
vision
conservation
aerial-imagery
drones
Model card Files
xet
 Community
ConservationDrones commited on
Commit
baf706f
·
verified ·
1 Parent(s): 038ca9f

Upload 5 files

Browse files
Files changed (6) hide show
  1. .gitattributes +1 -0
  2. config.json +119 -0
  3. conversion_checks.json +157 -0
  4. example_annotated.jpg +3 -0
  5. model.safetensors +3 -0
  6. preprocessor_config.json +25 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ example_annotated.jpg filter=lfs diff=lfs merge=lfs -text
config.json ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.0,
3
+ "activation_function": "silu",
4
+ "architectures": [
5
+ "RfDetrForObjectDetection"
6
+ ],
7
+ "attention_bias": true,
8
+ "attention_dropout": 0.0,
9
+ "auxiliary_loss": true,
10
+ "backbone_config": {
11
+ "apply_layernorm": true,
12
+ "attention_probs_dropout_prob": 0.0,
13
+ "drop_path_rate": 0.0,
14
+ "dtype": "float32",
15
+ "hidden_act": "gelu",
16
+ "hidden_dropout_prob": 0.0,
17
+ "hidden_size": 384,
18
+ "image_size": 1280,
19
+ "initializer_range": 0.02,
20
+ "layer_norm_eps": 1e-06,
21
+ "layerscale_value": 1.0,
22
+ "mlp_ratio": 4,
23
+ "model_type": "rf_detr_dinov2",
24
+ "num_attention_heads": 6,
25
+ "num_channels": 3,
26
+ "num_hidden_layers": 12,
27
+ "num_windows": 2,
28
+ "out_features": [
29
+ "stage3",
30
+ "stage6",
31
+ "stage9",
32
+ "stage12"
33
+ ],
34
+ "out_indices": [
35
+ 3,
36
+ 6,
37
+ 9,
38
+ 12
39
+ ],
40
+ "patch_size": 16,
41
+ "qkv_bias": true,
42
+ "reshape_hidden_states": true,
43
+ "stage_names": [
44
+ "stem",
45
+ "stage1",
46
+ "stage2",
47
+ "stage3",
48
+ "stage4",
49
+ "stage5",
50
+ "stage6",
51
+ "stage7",
52
+ "stage8",
53
+ "stage9",
54
+ "stage10",
55
+ "stage11",
56
+ "stage12"
57
+ ],
58
+ "use_mask_token": true,
59
+ "use_swiglu_ffn": false,
60
+ "window_block_indexes": [
61
+ 0,
62
+ 1,
63
+ 2,
64
+ 4,
65
+ 5,
66
+ 7,
67
+ 8,
68
+ 10,
69
+ 11
70
+ ]
71
+ },
72
+ "bbox_cost": 5,
73
+ "bbox_loss_coefficient": 5,
74
+ "c2f_num_blocks": 3,
75
+ "class_cost": 2,
76
+ "class_loss_coefficient": 1,
77
+ "d_model": 256,
78
+ "decoder_activation_function": "relu",
79
+ "decoder_cross_attention_heads": 16,
80
+ "decoder_ffn_dim": 2048,
81
+ "decoder_layers": 4,
82
+ "decoder_n_points": 2,
83
+ "decoder_self_attention_heads": 8,
84
+ "dice_loss_coefficient": 1,
85
+ "disable_custom_kernels": true,
86
+ "dropout": 0.1,
87
+ "dtype": "float32",
88
+ "eos_coefficient": 0.1,
89
+ "focal_alpha": 0.25,
90
+ "giou_cost": 2,
91
+ "giou_loss_coefficient": 2,
92
+ "group_detr": 13,
93
+ "hidden_expansion": 0.5,
94
+ "id2label": {
95
+ "0": "person",
96
+ "1": "animal",
97
+ "2": "vehicle",
98
+ "3": "unused_legacy_slot"
99
+ },
100
+ "init_std": 0.02,
101
+ "intermediate_size": 1024,
102
+ "label2id": {
103
+ "animal": 1,
104
+ "person": 0,
105
+ "unused_legacy_slot": 3,
106
+ "vehicle": 2
107
+ },
108
+ "layer_norm_eps": 1e-05,
109
+ "mask_class_loss_coefficient": 5.0,
110
+ "mask_dice_loss_coefficient": 5.0,
111
+ "mask_downsample_ratio": 4,
112
+ "mask_loss_coefficient": 1,
113
+ "mask_point_sample_ratio": 16,
114
+ "model_type": "rf_detr",
115
+ "num_feature_levels": 1,
116
+ "num_queries": 300,
117
+ "segmentation_head_activation_function": "gelu",
118
+ "transformers_version": "5.18.0"
119
+ }
conversion_checks.json ADDED
@@ -0,0 +1,157 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "source": "/Users/josh/code/boem-genai/DroneMegaDetector_RFDETRMedium.pth",
3
+ "source_sha256": "d729d9cd7ac3bc5c2f4442172035066496a10c8c748d25786d515ce1c16a2856",
4
+ "checkpoint_rfdetr_version": "1.6.0",
5
+ "reference_rfdetr_version": "1.6.0",
6
+ "transformers_version": "5.18.0",
7
+ "missing_keys": [],
8
+ "unexpected_keys": [],
9
+ "mismatched_keys": [],
10
+ "position_grid": 80,
11
+ "processor_resolution": 1280,
12
+ "labels": {
13
+ "0": "person",
14
+ "1": "animal",
15
+ "2": "vehicle",
16
+ "3": "unused_legacy_slot"
17
+ },
18
+ "checks": {
19
+ "224px_float32_backbone": {
20
+ "shape": [
21
+ 2,
22
+ 384,
23
+ 14,
24
+ 14
25
+ ],
26
+ "max_absolute_error": 3.4332275390625e-05,
27
+ "mean_absolute_error": 3.6743040254805237e-06,
28
+ "atol": 0.0002,
29
+ "rtol": 0.0002
30
+ },
31
+ "320px_logits": {
32
+ "shape": [
33
+ 2,
34
+ 300,
35
+ 4
36
+ ],
37
+ "max_absolute_error": 3.8130639561373414e-06,
38
+ "mean_absolute_error": 4.880644085079414e-07,
39
+ "atol": 0.0002,
40
+ "rtol": 0.0002
41
+ },
42
+ "320px_boxes": {
43
+ "shape": [
44
+ 2,
45
+ 300,
46
+ 4
47
+ ],
48
+ "max_absolute_error": 5.856459688174942e-07,
49
+ "mean_absolute_error": 1.6388501435938554e-08,
50
+ "atol": 0.0002,
51
+ "rtol": 0.0002
52
+ },
53
+ "320px_backbone": {
54
+ "shape": [
55
+ 2,
56
+ 384,
57
+ 20,
58
+ 20
59
+ ],
60
+ "max_absolute_error": 9.769962616701378e-14,
61
+ "mean_absolute_error": 7.394314978870708e-15,
62
+ "atol": 0.0002,
63
+ "rtol": 0.0002
64
+ },
65
+ "576px_logits": {
66
+ "shape": [
67
+ 1,
68
+ 300,
69
+ 4
70
+ ],
71
+ "max_absolute_error": 3.4019432293419527e-06,
72
+ "mean_absolute_error": 4.856440021735094e-07,
73
+ "atol": 0.0002,
74
+ "rtol": 0.0002
75
+ },
76
+ "576px_boxes": {
77
+ "shape": [
78
+ 1,
79
+ 300,
80
+ 4
81
+ ],
82
+ "max_absolute_error": 3.4133945525649523e-07,
83
+ "mean_absolute_error": 1.4561927350870893e-08,
84
+ "atol": 0.0002,
85
+ "rtol": 0.0002
86
+ },
87
+ "576px_backbone": {
88
+ "shape": [
89
+ 1,
90
+ 384,
91
+ 36,
92
+ 36
93
+ ],
94
+ "max_absolute_error": 9.514611321037592e-14,
95
+ "mean_absolute_error": 7.84452137747884e-15,
96
+ "atol": 0.0002,
97
+ "rtol": 0.0002
98
+ },
99
+ "reload_logits": {
100
+ "shape": [
101
+ 1,
102
+ 300,
103
+ 4
104
+ ],
105
+ "max_absolute_error": 0.0,
106
+ "mean_absolute_error": 0.0,
107
+ "atol": 0.0002,
108
+ "rtol": 0.0002
109
+ },
110
+ "reload_boxes": {
111
+ "shape": [
112
+ 1,
113
+ 300,
114
+ 4
115
+ ],
116
+ "max_absolute_error": 0.0,
117
+ "mean_absolute_error": 0.0,
118
+ "atol": 0.0002,
119
+ "rtol": 0.0002
120
+ },
121
+ "processor_reload": {
122
+ "shape": [
123
+ 1,
124
+ 3,
125
+ 1280,
126
+ 1280
127
+ ],
128
+ "max_absolute_error": 0.0,
129
+ "mean_absolute_error": 0.0,
130
+ "atol": 0,
131
+ "rtol": 0
132
+ },
133
+ "native_logits": {
134
+ "shape": [
135
+ 1,
136
+ 300,
137
+ 4
138
+ ],
139
+ "max_absolute_error": 1.9140028244102325e-05,
140
+ "mean_absolute_error": 7.988022616367551e-07,
141
+ "atol": 0.0002,
142
+ "rtol": 0.0002
143
+ },
144
+ "native_boxes": {
145
+ "shape": [
146
+ 1,
147
+ 300,
148
+ 4
149
+ ],
150
+ "max_absolute_error": 2.770148079811463e-06,
151
+ "mean_absolute_error": 4.032394474988902e-08,
152
+ "atol": 0.0002,
153
+ "rtol": 0.0002
154
+ }
155
+ },
156
+ "detector_reference_precision": "float64 (reduces top-k near-tie sensitivity)"
157
+ }
example_annotated.jpg ADDED

Git LFS Details

  • SHA256: 175adcb569a7ee3ecd58993d2e87159894643b6f42548fca540bec9bccc430cb
  • Pointer size: 132 Bytes
  • Size of remote file: 1.3 MB
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87cf3045c909b8eb2ce1cce39343a97806351c7eae5a6e5cb48f96fc0c20cf1b
3
+ size 141400024
preprocessor_config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_annotations": true,
3
+ "do_normalize": true,
4
+ "do_pad": true,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "format": "coco_detection",
8
+ "image_mean": [
9
+ 0.485,
10
+ 0.456,
11
+ 0.406
12
+ ],
13
+ "image_processor_type": "RfDetrImageProcessor",
14
+ "image_std": [
15
+ 0.229,
16
+ 0.224,
17
+ 0.225
18
+ ],
19
+ "resample": 2,
20
+ "rescale_factor": 0.00392156862745098,
21
+ "size": {
22
+ "height": 1280,
23
+ "width": 1280
24
+ }
25
+ }