Hugging Face's logo

hf-tiny-v2
/
tiny-random-InstructBlipForConditionalGeneration

Safetensors
instructblip
Model card Files
xet
 Community
hf-transformers-bot commited on
Commit
c0c25f9
·
verified ·
1 Parent(s): 31329be

Update tiny models for InstructBlipForConditionalGeneration

Browse files
Files changed (7) hide show
  1. config.json +76 -0
  2. generation_config.json +10 -0
  3. model.safetensors +3 -0
  4. preprocessor_config.json +27 -0
  5. tokenizer.json +0 -0
  6. tokenizer_config.json +15 -0
  7. video_preprocessor_config.json +29 -0
config.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "InstructBlipForConditionalGeneration"
4
+ ],
5
+ "dtype": "float32",
6
+ "image_token_index": 4,
7
+ "initializer_factor": 1.0,
8
+ "initializer_range": 0.02,
9
+ "model_type": "instructblip",
10
+ "num_query_tokens": 10,
11
+ "qformer_config": {
12
+ "attention_dropout": 0.1,
13
+ "attention_probs_dropout_prob": 0.1,
14
+ "bos_token_id": 0,
15
+ "cross_attention_frequency": 2,
16
+ "dropout": 0.1,
17
+ "encoder_hidden_size": 32,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 32,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 37,
23
+ "layer_norm_eps": 1e-12,
24
+ "max_position_embeddings": 512,
25
+ "model_type": "instructblip_qformer",
26
+ "num_attention_heads": 4,
27
+ "num_hidden_layers": 2,
28
+ "pad_token_id": 0,
29
+ "projection_dim": 32,
30
+ "vocab_size": 99
31
+ },
32
+ "text_config": {
33
+ "_remove_final_layer_norm": false,
34
+ "activation_function": "relu",
35
+ "attention_dropout": 0.1,
36
+ "bos_token_id": 0,
37
+ "do_layer_norm_before": true,
38
+ "dropout": 0.1,
39
+ "embed_dim": 16,
40
+ "enable_bias": true,
41
+ "eos_token_id": 1,
42
+ "ffn_dim": 4,
43
+ "hidden_size": 16,
44
+ "init_std": 0.02,
45
+ "layer_norm_elementwise_affine": true,
46
+ "layerdrop": 0.0,
47
+ "max_position_embeddings": 200,
48
+ "model_type": "opt",
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 2,
51
+ "pad_token_id": 0,
52
+ "tie_word_embeddings": true,
53
+ "use_cache": true,
54
+ "vocab_size": 32102,
55
+ "word_embed_proj_dim": 16
56
+ },
57
+ "transformers_version": "5.16.0.dev0",
58
+ "use_decoder_only_language_model": true,
59
+ "vision_config": {
60
+ "attention_dropout": 0.1,
61
+ "dropout": 0.1,
62
+ "hidden_act": "gelu",
63
+ "hidden_size": 32,
64
+ "image_size": 30,
65
+ "initializer_range": 1e-10,
66
+ "intermediate_size": 37,
67
+ "layer_norm_eps": 1e-06,
68
+ "model_type": "instructblip_vision_model",
69
+ "num_attention_heads": 4,
70
+ "num_channels": 3,
71
+ "num_hidden_layers": 2,
72
+ "patch_size": 2,
73
+ "projection_dim": 32,
74
+ "qkv_bias": true
75
+ }
76
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "output_attentions": false,
6
+ "output_hidden_states": false,
7
+ "pad_token_id": 0,
8
+ "transformers_version": "5.16.0.dev0",
9
+ "use_cache": true
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b53c9b4c2d6a77db1492bf7caf856b3ff0db60603c540805ce21448d74b447a0
3
+ size 2351512
preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 30,
4
+ "width": 30
5
+ },
6
+ "do_convert_rgb": true,
7
+ "do_normalize": true,
8
+ "do_rescale": true,
9
+ "do_resize": true,
10
+ "image_mean": [
11
+ 0.48145466,
12
+ 0.4578275,
13
+ 0.40821073
14
+ ],
15
+ "image_processor_type": "BlipImageProcessor",
16
+ "image_std": [
17
+ 0.26862954,
18
+ 0.26130258,
19
+ 0.27577711
20
+ ],
21
+ "resample": 3,
22
+ "rescale_factor": 0.00392156862745098,
23
+ "size": {
24
+ "height": 30,
25
+ "width": 30
26
+ }
27
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "clean_up_tokenization_spaces": true,
4
+ "eos_token": "</s>",
5
+ "extra_ids": 100,
6
+ "is_local": true,
7
+ "local_files_only": false,
8
+ "model_max_length": 512,
9
+ "pad_token": "<pad>",
10
+ "processor_class": "InstructBlipProcessor",
11
+ "sp_model_kwargs": {},
12
+ "tokenizer_class": "TokenizersBackend",
13
+ "truncation_side": "left",
14
+ "unk_token": "<unk>"
15
+ }
video_preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 30,
4
+ "width": 30
5
+ },
6
+ "do_convert_rgb": true,
7
+ "do_normalize": true,
8
+ "do_rescale": true,
9
+ "do_resize": true,
10
+ "do_sample_frames": false,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_std": [
17
+ 0.26862954,
18
+ 0.26130258,
19
+ 0.27577711
20
+ ],
21
+ "resample": 3,
22
+ "rescale_factor": 0.00392156862745098,
23
+ "return_metadata": false,
24
+ "size": {
25
+ "height": 30,
26
+ "width": 30
27
+ },
28
+ "video_processor_type": "InstructBlipVideoVideoProcessor"
29
+ }