add model
Browse files- config.json +36 -0
- pytorch_model.bin +3 -0
config.json
ADDED
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"Pix2SeqForConditionalGeneration"
|
4 |
+
],
|
5 |
+
"attention_probs_dropout_prob": 0.0,
|
6 |
+
"bos_token_id": 10,
|
7 |
+
"dec_proj_mode": "mlp",
|
8 |
+
"dim_decoder": 512,
|
9 |
+
"drop_path_rate": 0.1,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.0,
|
12 |
+
"hidden_size": 768,
|
13 |
+
"image_size": 640,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 3072,
|
16 |
+
"is_encoder_decoder": true,
|
17 |
+
"layer_norm_eps": 1e-12,
|
18 |
+
"max_position_embeddings": 512,
|
19 |
+
"model_type": "pix2seq",
|
20 |
+
"num_attention_heads_decoder": 16,
|
21 |
+
"num_attention_heads_encoder": 12,
|
22 |
+
"num_channels": 3,
|
23 |
+
"num_decoder_layers": 6,
|
24 |
+
"num_encoder_layers": 12,
|
25 |
+
"output_bias": true,
|
26 |
+
"patch_size": 16,
|
27 |
+
"positional_encoding": "sin_cos",
|
28 |
+
"positional_encoding_decoder": "learned",
|
29 |
+
"qkv_bias": true,
|
30 |
+
"tie_word_embeddings": false,
|
31 |
+
"torch_dtype": "float32",
|
32 |
+
"transformers_version": "4.19.0.dev0",
|
33 |
+
"use_cache": true,
|
34 |
+
"use_cls_token": false,
|
35 |
+
"vocab_size": 3000
|
36 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:18b84ea2f6be34def678472dd71ac858efea8abff841af65ebcd29d11c847e8b
|
3 |
+
size 466940829
|