Upload 3 files
Browse files- config.json +131 -0
- model.safetensors +3 -0
- preprocessor_config.json +27 -0
config.json
ADDED
@@ -0,0 +1,131 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_projection": false,
|
3 |
+
"architectures": [
|
4 |
+
"DPTForDepthEstimation"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.0,
|
7 |
+
"auxiliary_loss_weight": 0.4,
|
8 |
+
"backbone": null,
|
9 |
+
"backbone_config": {
|
10 |
+
"add_fpn": false,
|
11 |
+
"attention_probs_dropout_prob": 0.0,
|
12 |
+
"auxiliary_channels": 256,
|
13 |
+
"auxiliary_concat_input": false,
|
14 |
+
"auxiliary_loss_weight": 0.4,
|
15 |
+
"auxiliary_num_convs": 1,
|
16 |
+
"drop_path_rate": 0.1,
|
17 |
+
"hidden_act": "gelu",
|
18 |
+
"hidden_dropout_prob": 0.0,
|
19 |
+
"hidden_size": 1024,
|
20 |
+
"image_size": 384,
|
21 |
+
"initializer_range": 0.02,
|
22 |
+
"intermediate_size": 4096,
|
23 |
+
"layer_norm_eps": 1e-12,
|
24 |
+
"layer_scale_init_value": 0.1,
|
25 |
+
"model_type": "beit",
|
26 |
+
"num_attention_heads": 16,
|
27 |
+
"num_channels": 3,
|
28 |
+
"num_hidden_layers": 24,
|
29 |
+
"out_features": [
|
30 |
+
"stage6",
|
31 |
+
"stage12",
|
32 |
+
"stage18",
|
33 |
+
"stage24"
|
34 |
+
],
|
35 |
+
"out_indices": [
|
36 |
+
6,
|
37 |
+
12,
|
38 |
+
18,
|
39 |
+
24
|
40 |
+
],
|
41 |
+
"patch_size": 16,
|
42 |
+
"pool_scales": [
|
43 |
+
1,
|
44 |
+
2,
|
45 |
+
3,
|
46 |
+
6
|
47 |
+
],
|
48 |
+
"reshape_hidden_states": false,
|
49 |
+
"semantic_loss_ignore_index": 255,
|
50 |
+
"stage_names": [
|
51 |
+
"stem",
|
52 |
+
"stage1",
|
53 |
+
"stage2",
|
54 |
+
"stage3",
|
55 |
+
"stage4",
|
56 |
+
"stage5",
|
57 |
+
"stage6",
|
58 |
+
"stage7",
|
59 |
+
"stage8",
|
60 |
+
"stage9",
|
61 |
+
"stage10",
|
62 |
+
"stage11",
|
63 |
+
"stage12",
|
64 |
+
"stage13",
|
65 |
+
"stage14",
|
66 |
+
"stage15",
|
67 |
+
"stage16",
|
68 |
+
"stage17",
|
69 |
+
"stage18",
|
70 |
+
"stage19",
|
71 |
+
"stage20",
|
72 |
+
"stage21",
|
73 |
+
"stage22",
|
74 |
+
"stage23",
|
75 |
+
"stage24"
|
76 |
+
],
|
77 |
+
"use_absolute_position_embeddings": false,
|
78 |
+
"use_auxiliary_head": true,
|
79 |
+
"use_mask_token": false,
|
80 |
+
"use_mean_pooling": true,
|
81 |
+
"use_relative_position_bias": true,
|
82 |
+
"use_shared_relative_position_bias": false,
|
83 |
+
"vocab_size": 8192
|
84 |
+
},
|
85 |
+
"backbone_featmap_shape": null,
|
86 |
+
"backbone_kwargs": null,
|
87 |
+
"backbone_out_indices": null,
|
88 |
+
"fusion_hidden_size": 256,
|
89 |
+
"head_in_index": -1,
|
90 |
+
"hidden_act": "gelu",
|
91 |
+
"hidden_dropout_prob": 0.0,
|
92 |
+
"hidden_size": 768,
|
93 |
+
"image_size": 384,
|
94 |
+
"initializer_range": 0.02,
|
95 |
+
"intermediate_size": 3072,
|
96 |
+
"is_hybrid": false,
|
97 |
+
"layer_norm_eps": 1e-12,
|
98 |
+
"model_type": "dpt",
|
99 |
+
"neck_hidden_sizes": [
|
100 |
+
256,
|
101 |
+
512,
|
102 |
+
1024,
|
103 |
+
1024
|
104 |
+
],
|
105 |
+
"neck_ignore_stages": [],
|
106 |
+
"num_attention_heads": 12,
|
107 |
+
"num_channels": 3,
|
108 |
+
"num_hidden_layers": 12,
|
109 |
+
"output_attentions": false,
|
110 |
+
"patch_size": 16,
|
111 |
+
"pooler_act": "tanh",
|
112 |
+
"pooler_output_size": 768,
|
113 |
+
"qkv_bias": true,
|
114 |
+
"readout_type": "project",
|
115 |
+
"reassemble_factors": [
|
116 |
+
4,
|
117 |
+
2,
|
118 |
+
1,
|
119 |
+
0.5
|
120 |
+
],
|
121 |
+
"semantic_classifier_dropout": 0.1,
|
122 |
+
"semantic_loss_ignore_index": 255,
|
123 |
+
"torch_dtype": "float32",
|
124 |
+
"transformers_version": null,
|
125 |
+
"use_autobackbone": true,
|
126 |
+
"use_auxiliary_head": true,
|
127 |
+
"use_batch_norm_in_fusion_residual": false,
|
128 |
+
"use_bias_in_fusion_residual": null,
|
129 |
+
"use_pretrained_backbone": false,
|
130 |
+
"use_timm_backbone": false
|
131 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0d353badaf2cb08a05ab343cc70d1895c769ff1da34da4e2962c08dce8d4d1a1
|
3 |
+
size 1373309364
|
preprocessor_config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"do_pad": false,
|
4 |
+
"do_reduce_labels": false,
|
5 |
+
"do_rescale": true,
|
6 |
+
"do_resize": true,
|
7 |
+
"ensure_multiple_of": 32,
|
8 |
+
"image_mean": [
|
9 |
+
0.5,
|
10 |
+
0.5,
|
11 |
+
0.5
|
12 |
+
],
|
13 |
+
"image_processor_type": "DPTImageProcessor",
|
14 |
+
"image_std": [
|
15 |
+
0.5,
|
16 |
+
0.5,
|
17 |
+
0.5
|
18 |
+
],
|
19 |
+
"keep_aspect_ratio": false,
|
20 |
+
"resample": 3,
|
21 |
+
"rescale_factor": 0.00392156862745098,
|
22 |
+
"size": {
|
23 |
+
"height": 384,
|
24 |
+
"width": 384
|
25 |
+
},
|
26 |
+
"size_divisor": null
|
27 |
+
}
|