add model
Browse files- config.json +44 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/data/ckpt/converted/vilt_pretrain",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ViltForQuestionAnswering"
|
| 5 |
+
],
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"contrast_method": "none",
|
| 8 |
+
"distill": false,
|
| 9 |
+
"draw_false_text": 0,
|
| 10 |
+
"drop_rate": 0.1,
|
| 11 |
+
"eos_token_id": 2,
|
| 12 |
+
"freeze_patterns": null,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"ib_kl": false,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"itm_loss": 1,
|
| 17 |
+
"keep_ratio": 1,
|
| 18 |
+
"max_text_len": 40,
|
| 19 |
+
"merge_r": 0,
|
| 20 |
+
"merge_ratio": 0,
|
| 21 |
+
"merge_style": "tip",
|
| 22 |
+
"merge_text": 0,
|
| 23 |
+
"mlm_loss": 1,
|
| 24 |
+
"mlm_prob": 0.15,
|
| 25 |
+
"mlp_ratio": 4,
|
| 26 |
+
"model_type": "vilt",
|
| 27 |
+
"mpp_loss": 0,
|
| 28 |
+
"num_heads": 12,
|
| 29 |
+
"num_layers": 12,
|
| 30 |
+
"pad_token_id": 1,
|
| 31 |
+
"prune_layers": null,
|
| 32 |
+
"prune_method": "mlp_states",
|
| 33 |
+
"prune_r": 0,
|
| 34 |
+
"reduce_layers": null,
|
| 35 |
+
"sim_method": "first_head",
|
| 36 |
+
"token_types": 2,
|
| 37 |
+
"tokenizer": "bert-base-uncased",
|
| 38 |
+
"torch_dtype": "float32",
|
| 39 |
+
"transformers_version": "4.25.1",
|
| 40 |
+
"vit": "vit_base_patch32_384",
|
| 41 |
+
"vocab_size": 30522,
|
| 42 |
+
"vqa_label_size": 3129,
|
| 43 |
+
"whole_word_masking": false
|
| 44 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95fd12569ee159cbe4ed478bbf65c0688a7ecf69e909a9b3dd171f574c4a22e4
|
| 3 |
+
size 470407671
|