Upload model

config.json

@@ -0,0 +1,26 @@
+{
+  "architectures": [
+    "VitMemModel"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_vitmodel.ViTConfig",
+    "AutoModel": "modeling_vitmodel.VitMemModel"
+  },
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.30.1"
+}

configuration_vitmodel.py

@@ -0,0 +1,39 @@
+from transformers import PretrainedConfig
+
+class ViTConfig(PretrainedConfig):
+    model_type = "vit"
+
+    def __init__(
+        self,
+        hidden_size=768,
+        num_hidden_layers=12,
+        num_attention_heads=12,
+        intermediate_size=3072,
+        hidden_act="gelu",
+        hidden_dropout_prob=0.0,
+        attention_probs_dropout_prob=0.0,
+        initializer_range=0.02,
+        layer_norm_eps=1e-12,
+        image_size=224,
+        patch_size=16,
+        num_channels=3,
+        qkv_bias=True,
+        encoder_stride=16,
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.hidden_size = hidden_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.intermediate_size = intermediate_size
+        self.hidden_act = hidden_act
+        self.hidden_dropout_prob = hidden_dropout_prob
+        self.attention_probs_dropout_prob = attention_probs_dropout_prob
+        self.initializer_range = initializer_range
+        self.layer_norm_eps = layer_norm_eps
+        self.image_size = image_size
+        self.patch_size = patch_size
+        self.num_channels = num_channels
+        self.qkv_bias = qkv_bias
+        self.encoder_stride = encoder_stride

modeling_vitmodel.py

@@ -0,0 +1,16 @@
+import torch
+import timm
+from transformers import PreTrainedModel
+from .configuration_vitmodel import ViTConfig
+
+class VitMemModel(PreTrainedModel):
+    config_class = ViTConfig
+
+    def __init__(self, config: ViTConfig):
+        super().__init__(config)
+        self.model = timm.create_model("vit_base_patch16_224_miil", pretrained=False, num_classes=1)
+
+    def forward(self, tensor, labels=None):
+        vitfeat = self.model(tensor)
+        out = torch.sigmoid(vitfeat)
+        return out

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acb63c96d24c2ca7347982300c11360053a3f93b64dfa9b18057e0c516aef6dc
+size 343130109