GCancilla commited on
Commit
77b2f23
·
verified ·
1 Parent(s): 3d465ba

Upload 4 files

Browse files
config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "SamModel"
5
+ ],
6
+ "initializer_range": 0.02,
7
+ "mask_decoder_config": {
8
+ "attention_downsample_rate": 2,
9
+ "hidden_act": "relu",
10
+ "hidden_size": 256,
11
+ "iou_head_depth": 3,
12
+ "iou_head_hidden_dim": 256,
13
+ "layer_norm_eps": 1e-06,
14
+ "mlp_dim": 2048,
15
+ "model_type": "",
16
+ "num_attention_heads": 8,
17
+ "num_hidden_layers": 2,
18
+ "num_multimask_outputs": 3,
19
+ "torch_dtype": "float32"
20
+ },
21
+ "model_type": "sam",
22
+ "prompt_encoder_config": {
23
+ "hidden_act": "gelu",
24
+ "hidden_size": 256,
25
+ "image_embedding_size": 64,
26
+ "image_size": 1024,
27
+ "layer_norm_eps": 1e-06,
28
+ "mask_input_channels": 16,
29
+ "model_type": "",
30
+ "num_point_embeddings": 4,
31
+ "patch_size": 16,
32
+ "torch_dtype": "float32"
33
+ },
34
+ "torch_dtype": "float32",
35
+ "transformers_version": "4.52.0.dev0",
36
+ "vision_config": {
37
+ "attention_dropout": 0.0,
38
+ "global_attn_indexes": [
39
+ 2,
40
+ 5,
41
+ 8,
42
+ 11
43
+ ],
44
+ "hidden_act": "gelu",
45
+ "hidden_size": 168,
46
+ "image_size": 1024,
47
+ "initializer_range": 1e-10,
48
+ "layer_norm_eps": 1e-06,
49
+ "mlp_dim": 696,
50
+ "mlp_ratio": 4.0,
51
+ "model_type": "sam_vision_model",
52
+ "num_attention_heads": 12,
53
+ "num_channels": 3,
54
+ "num_hidden_layers": 12,
55
+ "num_pos_feats": 128,
56
+ "output_channels": 256,
57
+ "patch_size": 16,
58
+ "qkv_bias": true,
59
+ "torch_dtype": "float32",
60
+ "use_abs_pos": true,
61
+ "use_rel_pos": true,
62
+ "window_size": 14
63
+ }
64
+ }
onnx/prompt_encoder_mask_decoder.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4849ce851fc3466422b860064bff1d28c5aca1821a8514191941979117f52e1e
3
+ size 16515029
onnx/vision_encoder.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb2bce9618c71f9dfe8488558784d7e368d5093aa329bd1157849deddb28d730
3
+ size 28987339
preprocessor_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": true,
3
+ "do_normalize": true,
4
+ "do_pad": true,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.485,
9
+ 0.456,
10
+ 0.406
11
+ ],
12
+ "image_processor_type": "SamImageProcessor",
13
+ "image_std": [
14
+ 0.229,
15
+ 0.224,
16
+ 0.225
17
+ ],
18
+ "mask_pad_size": {
19
+ "height": 256,
20
+ "width": 256
21
+ },
22
+ "mask_size": {
23
+ "longest_edge": 256
24
+ },
25
+ "pad_size": {
26
+ "height": 1024,
27
+ "width": 1024
28
+ },
29
+ "processor_class": "SamProcessor",
30
+ "resample": 2,
31
+ "rescale_factor": 0.00392156862745098,
32
+ "size": {
33
+ "longest_edge": 1024
34
+ }
35
+ }