upload metaclip
Browse files- MetaCLIP/.DS_Store +0 -0
- MetaCLIP/metaclip-h14-fullcc2.5b/added_tokens.json +4 -0
- MetaCLIP/metaclip-h14-fullcc2.5b/config.json +26 -0
- MetaCLIP/metaclip-h14-fullcc2.5b/merges.txt +0 -0
- MetaCLIP/metaclip-h14-fullcc2.5b/preprocessor_config.json +28 -0
- MetaCLIP/metaclip-h14-fullcc2.5b/pytorch_model.bin +3 -0
- MetaCLIP/metaclip-h14-fullcc2.5b/special_tokens_map.json +6 -0
- MetaCLIP/metaclip-h14-fullcc2.5b/tokenizer_config.json +33 -0
- MetaCLIP/metaclip-h14-fullcc2.5b/vocab.json +0 -0
- MetaCLIP/metaclip-l14-fullcc2.5b/added_tokens.json +4 -0
- MetaCLIP/metaclip-l14-fullcc2.5b/config.json +25 -0
- MetaCLIP/metaclip-l14-fullcc2.5b/merges.txt +0 -0
- MetaCLIP/metaclip-l14-fullcc2.5b/preprocessor_config.json +28 -0
- MetaCLIP/metaclip-l14-fullcc2.5b/pytorch_model.bin +3 -0
- MetaCLIP/metaclip-l14-fullcc2.5b/special_tokens_map.json +6 -0
- MetaCLIP/metaclip-l14-fullcc2.5b/tokenizer_config.json +33 -0
- MetaCLIP/metaclip-l14-fullcc2.5b/vocab.json +0 -0
MetaCLIP/.DS_Store
ADDED
Binary file (6.15 kB). View file
|
|
MetaCLIP/metaclip-h14-fullcc2.5b/added_tokens.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<|endoftext|>": 49407,
|
3 |
+
"<|startoftext|>": 49406
|
4 |
+
}
|
MetaCLIP/metaclip-h14-fullcc2.5b/config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"CLIPModel"
|
4 |
+
],
|
5 |
+
"initializer_factor": 1.0,
|
6 |
+
"logit_scale_init_value": 2.6592,
|
7 |
+
"model_type": "clip",
|
8 |
+
"projection_dim": 1024,
|
9 |
+
"text_config": {
|
10 |
+
"hidden_size": 1024,
|
11 |
+
"intermediate_size": 4096,
|
12 |
+
"model_type": "clip_text_model",
|
13 |
+
"num_attention_heads": 16,
|
14 |
+
"num_hidden_layers": 24
|
15 |
+
},
|
16 |
+
"torch_dtype": "float32",
|
17 |
+
"transformers_version": "4.34.0",
|
18 |
+
"vision_config": {
|
19 |
+
"hidden_size": 1280,
|
20 |
+
"intermediate_size": 5120,
|
21 |
+
"model_type": "clip_vision_model",
|
22 |
+
"num_attention_heads": 16,
|
23 |
+
"num_hidden_layers": 32,
|
24 |
+
"patch_size": 14
|
25 |
+
}
|
26 |
+
}
|
MetaCLIP/metaclip-h14-fullcc2.5b/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
MetaCLIP/metaclip-h14-fullcc2.5b/preprocessor_config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"crop_size": {
|
3 |
+
"height": 224,
|
4 |
+
"width": 224
|
5 |
+
},
|
6 |
+
"do_center_crop": true,
|
7 |
+
"do_convert_rgb": true,
|
8 |
+
"do_normalize": true,
|
9 |
+
"do_rescale": true,
|
10 |
+
"do_resize": true,
|
11 |
+
"image_mean": [
|
12 |
+
0.48145466,
|
13 |
+
0.4578275,
|
14 |
+
0.40821073
|
15 |
+
],
|
16 |
+
"image_processor_type": "CLIPImageProcessor",
|
17 |
+
"image_std": [
|
18 |
+
0.26862954,
|
19 |
+
0.26130258,
|
20 |
+
0.27577711
|
21 |
+
],
|
22 |
+
"processor_class": "CLIPProcessor",
|
23 |
+
"resample": 3,
|
24 |
+
"rescale_factor": 0.00392156862745098,
|
25 |
+
"size": {
|
26 |
+
"shortest_edge": 224
|
27 |
+
}
|
28 |
+
}
|
MetaCLIP/metaclip-h14-fullcc2.5b/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:730842177578983a18d47ed6794ae10375ce8a23d53f6b5233e4d87c6b7fce70
|
3 |
+
size 1972533122
|
MetaCLIP/metaclip-h14-fullcc2.5b/special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|startoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
MetaCLIP/metaclip-h14-fullcc2.5b/tokenizer_config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"49406": {
|
5 |
+
"content": "<|startoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": false
|
11 |
+
},
|
12 |
+
"49407": {
|
13 |
+
"content": "<|endoftext|>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": false,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false,
|
18 |
+
"special": false
|
19 |
+
}
|
20 |
+
},
|
21 |
+
"additional_special_tokens": [],
|
22 |
+
"bos_token": "<|startoftext|>",
|
23 |
+
"clean_up_tokenization_spaces": true,
|
24 |
+
"do_lower_case": true,
|
25 |
+
"eos_token": "<|endoftext|>",
|
26 |
+
"errors": "replace",
|
27 |
+
"model_max_length": 77,
|
28 |
+
"pad_token": "<|endoftext|>",
|
29 |
+
"processor_class": "CLIPProcessor",
|
30 |
+
"tokenizer_class": "CLIPTokenizer",
|
31 |
+
"tokenizer_file": "/Users/georgebredis/.cache/huggingface/hub/models--openai--clip-vit-base-patch32/snapshots/e6a30b603a447e251fdaca1c3056b2a16cdfebeb/tokenizer.json",
|
32 |
+
"unk_token": "<|endoftext|>"
|
33 |
+
}
|
MetaCLIP/metaclip-h14-fullcc2.5b/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
MetaCLIP/metaclip-l14-fullcc2.5b/added_tokens.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<|endoftext|>": 49407,
|
3 |
+
"<|startoftext|>": 49406
|
4 |
+
}
|
MetaCLIP/metaclip-l14-fullcc2.5b/config.json
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"CLIPModel"
|
4 |
+
],
|
5 |
+
"initializer_factor": 1.0,
|
6 |
+
"logit_scale_init_value": 2.6592,
|
7 |
+
"model_type": "clip",
|
8 |
+
"projection_dim": 768,
|
9 |
+
"text_config": {
|
10 |
+
"hidden_size": 768,
|
11 |
+
"intermediate_size": 3072,
|
12 |
+
"model_type": "clip_text_model",
|
13 |
+
"num_attention_heads": 12
|
14 |
+
},
|
15 |
+
"torch_dtype": "float32",
|
16 |
+
"transformers_version": "4.34.0",
|
17 |
+
"vision_config": {
|
18 |
+
"hidden_size": 1024,
|
19 |
+
"intermediate_size": 4096,
|
20 |
+
"model_type": "clip_vision_model",
|
21 |
+
"num_attention_heads": 16,
|
22 |
+
"num_hidden_layers": 24,
|
23 |
+
"patch_size": 14
|
24 |
+
}
|
25 |
+
}
|
MetaCLIP/metaclip-l14-fullcc2.5b/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
MetaCLIP/metaclip-l14-fullcc2.5b/preprocessor_config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"crop_size": {
|
3 |
+
"height": 224,
|
4 |
+
"width": 224
|
5 |
+
},
|
6 |
+
"do_center_crop": true,
|
7 |
+
"do_convert_rgb": true,
|
8 |
+
"do_normalize": true,
|
9 |
+
"do_rescale": true,
|
10 |
+
"do_resize": true,
|
11 |
+
"image_mean": [
|
12 |
+
0.48145466,
|
13 |
+
0.4578275,
|
14 |
+
0.40821073
|
15 |
+
],
|
16 |
+
"image_processor_type": "CLIPImageProcessor",
|
17 |
+
"image_std": [
|
18 |
+
0.26862954,
|
19 |
+
0.26130258,
|
20 |
+
0.27577711
|
21 |
+
],
|
22 |
+
"processor_class": "CLIPProcessor",
|
23 |
+
"resample": 3,
|
24 |
+
"rescale_factor": 0.00392156862745098,
|
25 |
+
"size": {
|
26 |
+
"shortest_edge": 224
|
27 |
+
}
|
28 |
+
}
|
MetaCLIP/metaclip-l14-fullcc2.5b/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a812f538be2d1454c383286d0d4f94cab59f76de9e3049b16e4a870a6cfbc579
|
3 |
+
size 855436674
|
MetaCLIP/metaclip-l14-fullcc2.5b/special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|startoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
MetaCLIP/metaclip-l14-fullcc2.5b/tokenizer_config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"49406": {
|
5 |
+
"content": "<|startoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": false
|
11 |
+
},
|
12 |
+
"49407": {
|
13 |
+
"content": "<|endoftext|>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": false,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false,
|
18 |
+
"special": false
|
19 |
+
}
|
20 |
+
},
|
21 |
+
"additional_special_tokens": [],
|
22 |
+
"bos_token": "<|startoftext|>",
|
23 |
+
"clean_up_tokenization_spaces": true,
|
24 |
+
"do_lower_case": true,
|
25 |
+
"eos_token": "<|endoftext|>",
|
26 |
+
"errors": "replace",
|
27 |
+
"model_max_length": 77,
|
28 |
+
"pad_token": "<|endoftext|>",
|
29 |
+
"processor_class": "CLIPProcessor",
|
30 |
+
"tokenizer_class": "CLIPTokenizer",
|
31 |
+
"tokenizer_file": "/Users/georgebredis/.cache/huggingface/hub/models--openai--clip-vit-base-patch32/snapshots/e6a30b603a447e251fdaca1c3056b2a16cdfebeb/tokenizer.json",
|
32 |
+
"unk_token": "<|endoftext|>"
|
33 |
+
}
|
MetaCLIP/metaclip-l14-fullcc2.5b/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|