Upload 5 files
Browse filesTVP model for temporal video grounding
- config.json +47 -0
- preprocessor_config.json +13 -0
- pytorch_model.bin +3 -0
- tokenizer.json +0 -0
- vocab.txt +0 -0
config.json
ADDED
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"model_type": "tvp",
|
3 |
+
"alpha": 1.0,
|
4 |
+
"beta": 0.1,
|
5 |
+
"vp_type": "framepad",
|
6 |
+
"vp_apply": "replace",
|
7 |
+
"max_img_size": 448,
|
8 |
+
"pad_size": 96,
|
9 |
+
"num_frm": 48,
|
10 |
+
"attention_probs_dropout_prob": 0.1,
|
11 |
+
"hidden_size": 768,
|
12 |
+
"classifier": "mlp",
|
13 |
+
"classifier_dropout": null,
|
14 |
+
"cls_hidden_scale": 2,
|
15 |
+
"hidden_act": "gelu",
|
16 |
+
"hidden_dropout_prob": 0.1,
|
17 |
+
"initializer_range": 0.02,
|
18 |
+
"intermediate_size": 3072,
|
19 |
+
"layer_norm_eps": 1e-12,
|
20 |
+
"max_grid_col_position_embeddings": 100,
|
21 |
+
"max_grid_row_position_embeddings": 100,
|
22 |
+
"max_position_embeddings": 512,
|
23 |
+
"num_attention_heads": 12,
|
24 |
+
"num_hidden_layers": 12,
|
25 |
+
"pad_token_id": 0,
|
26 |
+
"position_embedding_type": "absolute",
|
27 |
+
"transformers_version": "4.29.2",
|
28 |
+
"type_vocab_size": 2,
|
29 |
+
"use_cache": true,
|
30 |
+
"vocab_size": 30522,
|
31 |
+
"max_text_length": 100,
|
32 |
+
"vision_config": {
|
33 |
+
"model_type": "tvp_vision_model",
|
34 |
+
"input_format": "BGR",
|
35 |
+
"features": ["res5"],
|
36 |
+
"resnets_depth": 50,
|
37 |
+
"resnets_num_groups": 1,
|
38 |
+
"resnets_width_per_group": 64,
|
39 |
+
"resnets_stem_input_channels": 3,
|
40 |
+
"resnets_stem_out_channels": 64,
|
41 |
+
"resnets_res_out_channels": 256,
|
42 |
+
"resnets_res_dilation": 1,
|
43 |
+
"backbone_freeze_at": 2,
|
44 |
+
"grid_encoder_conv_input_size": 2048,
|
45 |
+
"grid_encoder_conv_output_size": 768
|
46 |
+
}
|
47 |
+
}
|
preprocessor_config.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_center_crop": false,
|
3 |
+
"do_normalize": true,
|
4 |
+
"do_resize": true,
|
5 |
+
"do_rescale": false,
|
6 |
+
"do_padding": true,
|
7 |
+
"image_mean": [8.2381, 7.3115, 6.6981],
|
8 |
+
"image_std": [9.6335, 9.0659, 8.7213],
|
9 |
+
"processor_class": "TVPProcessor",
|
10 |
+
"max_size": 448,
|
11 |
+
"padding_size": {"height": 448, "width": 448},
|
12 |
+
"tokenizer": "bert-base-uncased"
|
13 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fcae30a6c715639d1f5d4255a53ce0b479d058f96b3d2f7c554a1fa78410289a
|
3 |
+
size 673739769
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|