initial models

README.md

@@ -0,0 +1,42 @@
+---
+tags:
+- image-classification
+license: cc-by-nc-4.0
+---
+# Model card
+
+A Swin-v2 based species classification model.
+Original model: timm/swinv2_tiny_window16_256.ms_in1k
+
+## Model Details
+- **Model Type:** Animal classification
+- **Model Stats:**
+ - Image size: 256 x 256
+
+## Model Usage
+### Image Embeddings
+```python
+
+import timm
+import torch
+import torchvision.transforms as T
+
+from PIL import Image
+from urllib.request import urlopen
+
+model = timm.create_model("hf-hub:cermakvo/wildintel-species-classifier-swin-t", pretrained=True)
+model = model.eval()
+
+transforms = T.Compose([T.Resize(256), 
+                  T.ToTensor(), 
+                  T.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])]) 
+
+img = Image.open(urlopen(
+    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
+))
+
+output = model(transforms(img).unsqueeze(0))  # output is (batch_size, num_features) shaped tensor
+# output is a (1, num_features) shaped tensor
+```
+
+

config.json

@@ -0,0 +1,36 @@
+{
+  "architecture": "swinv2_tiny_window16_256",
+  "num_classes": 14,
+  "num_features": 768,
+  "global_pool": "avg",
+  "pretrained_cfg": {
+    "custom_load": false,
+    "input_size": [
+      3,
+      256,
+      256
+    ],
+    "fixed_input_size": true,
+    "interpolation": "bicubic",
+    "crop_pct": 0.9,
+    "crop_mode": "center",
+    "mean": [
+      0.485,
+      0.456,
+      0.406
+    ],
+    "std": [
+      0.229,
+      0.224,
+      0.225
+    ],
+    "num_classes": 1000,
+    "pool_size": [
+      8,
+      8
+    ],
+    "first_conv": "patch_embed.proj",
+    "classifier": "head.fc",
+    "license": "mit"
+  }
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be001670954070cbc1f8b5b1b00db295ced3fd5789bb1193bd6111a568ccabb4
+size 110424574