remove BEiT option

app.py

@@ -329,7 +329,7 @@ with gr.Blocks(theme='sudeepshouche/minimalist') as demo:
                 
             with gr.Column():
                 model_name = gr.Dropdown(
-                    ["Mummified 170", "Rock 170","Fossils BEiT","Fossils 142"],
+                    ["Mummified 170", "Rock 170","Fossils 142"],#"Fossils BEiT" removed
                     multiselect=False,
                     value="Fossils 142", # default option
                     label="Model",

closest_sample.py

@@ -25,7 +25,7 @@ embedding_fossils = np.load('dataset/embedding_fossils_170_finer.npy')
 
 fossils_pd= pd.read_csv('fossils_paths.csv')
 
-def pca_distance(pca,sample,embedding,top_k):
+def pca_distance(pca,sample,embedding,top_k): 
     """
     Args:
         pca:fitted PCA model
@@ -38,7 +38,7 @@ def pca_distance(pca,sample,embedding,top_k):
     all = pca.transform(embedding[:,-1])
     distances = np.linalg.norm(all - s, axis=1)
     sorted_indices = np.argsort(distances)
-    filtered_indices = sorted_indices[sorted_indices<=2852]
+    filtered_indices = sorted_indices[sorted_indices<=2852] # exclude general fossils, keep florissant only.
     top_indices = np.concatenate([filtered_indices[:2], filtered_indices[3:top_k+1]])
     return top_indices
 

explanations.py

@@ -128,8 +128,10 @@ def explain(model, input_image,explain_method,nb_samples,size=600, n_classes=171
             phi = np.abs(explainer(X, Y))[0]
             if len(phi.shape) == 3:
                 phi = np.mean(phi, -1)
+            #apply Gaussian smoothing
+            phi_smoothed = cv2.GaussianBlur(phi, (5, 5), sigmaX=1.0, sigmaY=1.0)
             show(X[0],output_size = size)
-            show(phi, output_size = size,p=1, alpha=0.2, )
+            show(phi_smoothed, output_size = size,p=1, alpha=0.2)
             # show(X[0][:,size_repetitions:2*size_repetitions,:])
             # show(phi[:,size_repetitions:2*size_repetitions], p=1, alpha=0.4)
             plt.savefig(f'phi_{e}{i}.png')