Update README.md

README.md

@@ -5,6 +5,7 @@ tags:
 - sklearn
 - skops
 - tabular-classification
+- visual emb-gam
 ---
 
 # Model description
@@ -71,9 +72,40 @@ Use the code below to get started with the model.
 <summary> Click to expand </summary>
 
 ```python
-import pickle 
-with open('model.pkl', 'rb') as file: 
-    clf = pickle.load(file)
+from PIL import Image
+from skops import hub_utils
+import torch
+from transformers import AutoFeatureExtractor, AutoModel
+import pickle
+import os
+
+# load embedding model
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+feature_extractor = AutoFeatureExtractor.from_pretrained('microsoft/resnet-50')
+model = AutoModel.from_pretrained('microsoft/resnet-50').eval().to(device)
+
+# load logistic regression
+os.mkdir('emb-gam-resnet')
+hub_utils.download(repo_id='Ramos-Ramos/emb-gam-resnet', dst='emb-gam-resnet')
+
+with open('emb-gam-resnet/model.pkl', 'rb') as file: 
+  logistic_regression = pickle.load(file)
+    
+# load image
+img = Image.open('examples/english_springer.png')
+
+# preprocess image
+inputs = {k: v.to(device) for k, v in feature_extractor(img, return_tensors='pt').items()}
+
+# extract patch embeddings
+with torch.no_grad():
+  patch_embeddings = rearrange(model(**inputs).last_hidden_state, 'b d h w -> b (h w) d').cpu()
+
+# classify
+pred = logistic_regression.predict(patch_embeddings.sum(dim=0, keepdim=True))
+
+# get patch contributions
+patch_contributions = logistic_regression.coef_ @ patch_embeddings.T.numpy()
 ```
 
 </details>