npvinHnivqn commited on
Commit
e4e8309
1 Parent(s): 8b653f1

Upload SepeT5ForVQA

Browse files
Files changed (3) hide show
  1. config.json +47 -0
  2. generation_config.json +4 -0
  3. model.safetensors +3 -0
config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "T5ModelConfig": {
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 3072,
7
+ "d_model": 768,
8
+ "decoder_start_token_id": 0,
9
+ "model_type": "t5",
10
+ "n_positions": 512,
11
+ "num_decoder_layers": 12,
12
+ "num_heads": 12,
13
+ "num_layers": 12,
14
+ "output_past": true,
15
+ "torch_dtype": "float32",
16
+ "vocab_size": 36096
17
+ },
18
+ "TextQformConfig": {
19
+ "encoder_hidden_size": 768,
20
+ "hidden_size": 768,
21
+ "intermediate_size": 2048,
22
+ "model_type": "blip_2_qformer",
23
+ "num_hidden_layers": 6
24
+ },
25
+ "VisualPerceiverConfig": {
26
+ "d_latents": 768,
27
+ "model_type": "perceiver",
28
+ "num_latents": 100,
29
+ "num_self_attends_per_block": 16
30
+ },
31
+ "VisualQformConfig": {
32
+ "encoder_hidden_size": 768,
33
+ "hidden_size": 768,
34
+ "intermediate_size": 2048,
35
+ "model_type": "blip_2_qformer",
36
+ "num_hidden_layers": 6
37
+ },
38
+ "architectures": [
39
+ "SepeT5ForVQA"
40
+ ],
41
+ "continue_training": true,
42
+ "from_pretrained_t5": "VietAI/vit5-base",
43
+ "model_type": "qformt5",
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.36.0",
46
+ "visual_dmodel": 768
47
+ }
generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.36.0"
4
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2d9be1d3061f24b69925f74337520bc3aaa6d9a8b9585aa9865c3daf3419e08
3
+ size 1466806624