Hugging Face's logo

yuchenxie
/
Arlow-Vision-Encoder

Text Generation
Transformers
Safetensors
English
arlow
Model card Files
xet
 Community
yuchenxie commited on
Commit
49d4668
·
verified ·
1 Parent(s): ed8cd40

Upload folder using huggingface_hub

Browse files
Files changed (5) hide show
  1. config.json +36 -0
  2. image_processor/preprocessor_config.json +32 -0
  3. model.safetensors +3 -0
  4. processor_config.json +15 -0
  5. video_processor/video_preprocessor_config.json +34 -0
config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ArlowVLVisionModel"
4
+ ],
5
+ "deepstack_visual_indexes": [
6
+ 12,
7
+ 24,
8
+ 44
9
+ ],
10
+ "deformable_attention_strength": 4.0,
11
+ "deformable_attention_window": 0.25,
12
+ "depth": 48,
13
+ "dtype": "float16",
14
+ "embed_dim": 1536,
15
+ "hidden_act": "gelu_pytorch_tanh",
16
+ "hidden_size": 3072,
17
+ "in_channels": 3,
18
+ "initializer_range": 0.02,
19
+ "max_position_embeddings": 32768,
20
+ "mlp_ratio": 4,
21
+ "model_type": "arlow",
22
+ "mrope_sections": [
23
+ 21,
24
+ 21,
25
+ 22
26
+ ],
27
+ "num_attention_heads": 24,
28
+ "num_heads": 24,
29
+ "patch_size": 14,
30
+ "spatial_merge_size": 2,
31
+ "temporal_patch_size": 2,
32
+ "token_pruning_ratio": 0.0,
33
+ "transformers_version": "5.3.0.dev0",
34
+ "use_deformable_attention": true,
35
+ "use_progressive_patches": true
36
+ }
image_processor/preprocessor_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": true,
3
+ "do_normalize": true,
4
+ "do_pan_and_scan": false,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.5,
9
+ 0.5,
10
+ 0.5
11
+ ],
12
+ "image_processor_type": "ArlowImageProcessor",
13
+ "image_std": [
14
+ 0.5,
15
+ 0.5,
16
+ 0.5
17
+ ],
18
+ "max_pixels": 1003520,
19
+ "merge_size": 2,
20
+ "min_pixels": 3136,
21
+ "pan_and_scan_max_num_crops": 4,
22
+ "pan_and_scan_min_crop_size": 256,
23
+ "pan_and_scan_min_ratio_to_activate": 1.6,
24
+ "patch_size": 14,
25
+ "resample": 3,
26
+ "rescale_factor": 0.00392156862745098,
27
+ "size": {
28
+ "longest_edge": 1003520,
29
+ "shortest_edge": 3136
30
+ },
31
+ "temporal_patch_size": 2
32
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66839f7e3f1171bada1d1194620be49d3d83d217d0ec1d37c4c950b921a2603a
3
+ size 3181374712
processor_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "processor_class": "ArlowProcessor",
3
+ "image_processor_class": "ArlowImageProcessor",
4
+ "image_processor_subfolder": "image_processor",
5
+ "supports_multi_image": true,
6
+ "supports_video": true,
7
+ "patch_size": 14,
8
+ "temporal_patch_size": 2,
9
+ "merge_size": 2,
10
+ "schema_version": 1,
11
+ "video_processor_class": "ArlowVideoProcessor",
12
+ "video_processor_subfolder": "video_processor",
13
+ "max_frames": 96,
14
+ "timestamp_alignment": false
15
+ }
video_processor/video_preprocessor_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_last_selected_frame_indices": [],
3
+ "do_convert_rgb": true,
4
+ "do_normalize": true,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "do_sample_frames": true,
8
+ "image_mean": [
9
+ 0.5,
10
+ 0.5,
11
+ 0.5
12
+ ],
13
+ "image_std": [
14
+ 0.5,
15
+ 0.5,
16
+ 0.5
17
+ ],
18
+ "max_frames": 96,
19
+ "max_volume": 786432,
20
+ "merge_size": 2,
21
+ "min_frames": 4,
22
+ "motion_threshold": 0.1,
23
+ "patch_size": 14,
24
+ "resample": 3,
25
+ "rescale_factor": 0.00392156862745098,
26
+ "return_metadata": false,
27
+ "sample_strategy": "uniform",
28
+ "size": {
29
+ "longest_edge": 786432,
30
+ "shortest_edge": 131072
31
+ },
32
+ "temporal_patch_size": 2,
33
+ "video_processor_type": "ArlowVideoProcessor"
34
+ }