Upload ViTSiTForPreTraining
Browse files- config.json +23 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ViTSiTForPreTraining"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.0,
|
| 6 |
+
"hidden_act": "gelu",
|
| 7 |
+
"hidden_dropout_prob": 0.0,
|
| 8 |
+
"hidden_size": 384,
|
| 9 |
+
"image_size": 224,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 1536,
|
| 12 |
+
"layer_norm_eps": 1e-12,
|
| 13 |
+
"mask_ratio": 0.75,
|
| 14 |
+
"model_type": "vit_sit",
|
| 15 |
+
"num_attention_heads": 6,
|
| 16 |
+
"num_channels": 3,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"out_dim": 256,
|
| 19 |
+
"patch_size": 16,
|
| 20 |
+
"qkv_bias": true,
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.36.2"
|
| 23 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0112e216d75706bdc938c0f52fe691e0708624ee19d1ff1c300cdfd1fc9e1a73
|
| 3 |
+
size 167403452
|