refactored madhura's task 2 code, some more fixes

Phase 2/task_0.ipynb

@@ -137,9 +137,7 @@
     "    str(input(\"Enter feature model - one of \" + str(list(valid_feature_models.keys()))))\n",
     "]\n",
     "\n",
-    "selected_distance_measure = valid_distance_measures[\n",
-    "    str(input(\"Enter distance measure - one of \" + str(list(valid_distance_measures.keys()))))\n",
-    "]\n",
+    "selected_distance_measure = feature_distance_matches[selected_feature_model]\n",
     "\n",
     "if selected_image_id == -1:\n",
     "    show_similar_images_for_image(\n",

Phase 2/utils.py

@@ -291,7 +291,6 @@ def get_all_fd(image_id, given_image=None, given_label=None):
     else:
         img, label = dataset[image_id]
     img_shape = np.array(img).shape
-    print(img_shape)
     if img_shape[0] >= 3:
         true_channels = 3
     else:
@@ -357,7 +356,7 @@ valid_feature_models = {
     "avgpool": "avgpool_fd",
     "layer3": "layer3_fd",
     "fc": "fc_fd",
-    "resnet": "resnet_fd"
+    "resnet": "resnet_fd",
 }
 valid_distance_measures = {
     "euclidean": euclidean_distance_measure,
@@ -380,7 +379,7 @@ def show_similar_images_for_image(
     target_image=None,
     target_label=None,
     k=10,
-    feature_model="fc",
+    feature_model="fc_fd",
     distance_measure=pearson_distance_measure,
     save_plots=False,
 ):
@@ -509,7 +508,7 @@ def show_similar_images_for_image(
 
         if save_plots:
             plt.savefig(
-                f"Plots/Image_{target_image_id}_{feature_model}_{distance_measure.__name__}_k{k}.png"
+                f"Plots/Image_{target_image_id}_{feature_model}_{distance_measure.__name__}_{k}_images.png"
             )
         plt.show()
 
@@ -534,7 +533,7 @@ def show_similar_images_for_label(
     fd_collection,
     target_label,
     k=10,
-    feature_model="fc",
+    feature_model="fc_fd",
     distance_measure=pearson_distance_measure,
     save_plots=False,
 ):
@@ -594,7 +593,7 @@ def show_similar_images_for_label(
 
     if save_plots:
         plt.savefig(
-            f"Plots/Label_{target_label}_{feature_model}_{distance_measure.__name__}_k{k}.png"
+            f"Plots/Label_{target_label}_{feature_model}_{distance_measure.__name__}_{k}_images.png"
         )
     plt.show()
 
@@ -605,10 +604,18 @@ def show_similar_labels_for_image(
     target_image=None,
     target_label=None,
     k=10,
-    feature_model="fc",
+    feature_model="fc_fd",
     distance_measure=pearson_distance_measure,
     save_plots=False,
 ):
+    assert (
+        feature_model in valid_feature_models.values()
+    ), "feature_model should be one of " + str(valid_feature_models.keys())
+
+    assert (
+        distance_measure in valid_distance_measures.values()
+    ), "distance_measure should be one of " + str(list(valid_distance_measures.keys()))
+
     # if target from dataset
     if target_image_id != -1:
         print(
@@ -619,14 +626,14 @@ def show_similar_labels_for_image(
 
         # store target_image itself
         min_dists = {target_image_id: 0}
-        
+
         if target_image_id % 2 == 0:
             # Get target image's feature descriptors from database
             target_image = fd_collection.find_one({"image_id": target_image_id})
         else:
             # Calculate target image's feature descriptors
             target_image = get_all_fd(target_image_id)
-        
+
         target_image_fd = target_image[feature_model]
         target_label = target_image["true_label"]
 
@@ -644,62 +651,58 @@ def show_similar_labels_for_image(
         target_image_fd = np.array(target_image_fds[feature_model])
 
     label_dict = {target_image_id: target_label}
-    
-    target_image_fd = np.array(target_image[feature_model + "_fd"])
 
-    assert (
-        feature_model in valid_feature_models
-    ), "feature_model should be one of " + str(valid_feature_models)
-
-    assert (
-        distance_measure in valid_distance_measures.values()
-    ), "distance_measure should be one of " + str(list(valid_distance_measures.keys()))
-
-    # only RGB for non RGB images
-    if feature_model != "hog":
-        all_images = fd_collection.find({"true_channels": 3})
-    else:
-        all_images = fd_collection.find()
+    target_image_fd = np.array(target_image[feature_model])
 
+    all_images = fd_collection.find({})
     for cur_img in all_images:
         cur_img_id = cur_img["image_id"]
         # skip target itself
         if cur_img_id == target_image_id:
             continue
-        cur_img_fd = np.array(cur_img[feature_model + "_fd"])
+        cur_img_fd = np.array(cur_img[feature_model])
         cur_dist = distance_measure(
             cur_img_fd,
             target_image_fd,
         )
-        cursor = fd_collection.find({"image_id": cur_img_id})
-        label=cursor[0]["true_label"]
+        cur_label = cur_img["true_label"]
 
         # store first k images irrespective of distance (so that we store no more than k minimum distances)
-        if len(min_dists) < k + 1 and label not in label_dict.values():
+        if len(min_dists) < k + 1 and cur_label not in label_dict.values():
             min_dists[cur_img_id] = cur_dist
-            label_dict[cur_img_id] = label
+            label_dict[cur_img_id] = cur_label
 
         # if lower distance:
-        elif cur_dist < max(min_dists.values()) and label not in label_dict.values():
+        elif (
+            cur_dist < max(min_dists.values()) and cur_label not in label_dict.values()
+        ):
             # add to min_dists
             min_dists.update({cur_img_id: cur_dist})
-            label_dict.update({cur_img_id: label})
-            # remove greatest distance by index
-            pop_key=max(min_dists, key=min_dists.get)
+            label_dict.update({cur_img_id: cur_label})
+            # remove label with greatest distance by index
+            pop_key = max(min_dists, key=min_dists.get)
             min_dists.pop(pop_key)
             label_dict.pop(pop_key)
 
     min_dists = dict(sorted(min_dists.items(), key=lambda item: item[1]))
 
-    for image_id in min_dists.keys():
-        if image_id==target_image_id:
+    fig, axs = plt.subplots(1, k, figsize=(48, 12))
+    for idx, image_id in enumerate(min_dists.keys()):
+        if image_id == target_image_id:
             continue
         else:
-            print("Label: ", label_dict[image_id], "; distance: ", min_dists[image_id])
             sample_image, sample_label = dataset[image_id]
-            plt.imshow(transforms.ToPILImage()(sample_image))
-            plt.show()
-            
+            axs[idx-1].imshow(transforms.ToPILImage()(sample_image))
+            axs[idx-1].set_title(
+                f"Label: {label_dict[image_id]}; Distance: {min_dists[image_id]}"
+            )
+        axs[idx-1].axis("off")
+
+    if save_plots:
+        plt.savefig(
+            f"Plots/Image_{target_image_id}_{feature_model}_{distance_measure.__name__}_{k}_labels.png"
+        )
+    plt.show()
 
 
 valid_dim_reduction_methods = {
@@ -829,7 +832,7 @@ def extract_latent_semantics(
 
     all_images = list(fd_collection.find())
     feature_ids = [img["image_id"] for img in all_images]
-    
+
     top_img_str = ""
     if top_images is not None:
         top_img_str = f" (showing only top {top_images} image-weight pairs for each latent semantic)"
@@ -838,16 +841,16 @@ def extract_latent_semantics(
     if sim_matrix is not None:
         feature_vectors = sim_matrix
         print(
-            "Applying {} on the {} space to get {} latent semantics{}...".format(
-                dim_reduction_method, feature_model, k, top_img_str
+            "Applying {} on the given similarity matrix to get {} latent semantics{}...".format(
+                dim_reduction_method, k, top_img_str
             )
         )
     # else take feature space from database
     else:
-        feature_vectors = np.array([img[feature_model] for img in all_images])
+        feature_vectors = np.array([np.array(img[feature_model]).flatten() for img in all_images])
         print(
-            "Applying {} on the given similarity matrix to get {} latent semantics{}...".format(
-                dim_reduction_method, k, top_img_str
+            "Applying {} on the {} space to get {} latent semantics{}...".format(
+                dim_reduction_method, feature_model, k, top_img_str
             )
         )