Repository clean up for thesis

.gitignore

@@ -13,4 +13,4 @@ __pycache__
 *.aux
 *.lof
 *.log
-*.lot
+*.lot

src/data_set_up.py

@@ -2,7 +2,6 @@
 from collections import defaultdict
 import torch
 from torch.utils.data import Dataset
-import cv2
 import numpy as np
 from PIL import Image, ImageDraw
 

src/helpers.py

@@ -1,11 +1,7 @@
 import numpy as np
 import torch, statistics
-from sklearn.metrics import precision_recall_curve, average_precision_score
 import torch
 from torch.utils.data import DataLoader
-from torchvision.transforms import Compose, ToTensor, Normalize
-from torchvision import transforms
-from sklearn.metrics import confusion_matrix
 import pandas as pd
 from datetime import datetime
 
@@ -69,10 +65,6 @@ def train(model, train_loader, device, anchors, optimizer, num_epochs, dataset):
         # Save model info every 25 epochs to check progress with evaluation
         if epoch % 25 == 0 and epoch != 0: torch.save(model, 'results/model_{}_{}.pt'.format(dataset, epoch)) # type: ignore
         print('For the epoch:{} the loss is: {}, the precision is: {}, the recall is: {}, the mAP[0.5] is: {} and the mAP[0.5-0.95] is: {}'.format(loss, map['precision'], map['recall'], map['mAP'], mapCOCO['mAP']))
-
-
-
-
 
 
 def evaluate_model(model, data_loader, device, anchors):
@@ -107,4 +99,3 @@ def evaluate_model(model, data_loader, device, anchors):
         avg_loss = total_loss / len(data_loader)
 
     return avg_loss
-

test/test_extendedmask2former.py

@@ -16,51 +16,51 @@
 import torch.nn.functional as F
 from src.helpers import train, evaluate_model
 
-# # ------------------------------------------------------------------------------------------------------------------------------------
-# # Step 1: Load and preprocess the image
-# def load_image(image_path):
-#     image = Image.open(image_path).convert('RGB')
-#     transform = transforms.Compose([
-#         transforms.Resize((600, 600)),  # Resize to the input size expected by EfficientNet
-#         transforms.ToTensor(),
-#         transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),  # Normalization for EfficientNet
-#     ])
-#     return transform(image).unsqueeze(0)  # type: ignore # Add batch dimension
-
-# image = load_image('/Users/stamatiosorphanos/Documents/MCs_Thesis/SOD_Thesis/docs/Extended_Mask2Former/1.jpg')
-
-
-# # Step 2: Initialise the Mask2Former Model 
-# model = ExtendedMask2Former(num_classes=100, hidden_dim=256, num_queries=100, nheads=8, dim_feedforward=2048, dec_layers=1, mask_dim=256)
-
-# # Step 3: Count the number of parameters
-# total_params = sum(param.numel() for param in model.parameters())
-# print('The total number of Extended Mask2Former parameters are: ', total_params)
-# model.eval()  # Set the model to evaluation mode
-
-# # Step 4: Pass the image through the model
-# with torch.no_grad():
-#     output = model(image)
-
-# # Step 5: Visualize the feature maps
-# print('\nThe predicted classes are the following:\n')
-# print(output['pred_logits'])
+# ------------------------------------------------------------------------------------------------------------------------------------
+# Step 1: Load and preprocess the image
+def load_image(image_path):
+    image = Image.open(image_path).convert('RGB')
+    transform = transforms.Compose([
+        transforms.Resize((600, 600)),  # Resize to the input size expected by EfficientNet
+        transforms.ToTensor(),
+        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),  # Normalization for EfficientNet
+    ])
+    return transform(image).unsqueeze(0)  # type: ignore # Add batch dimension
+
+image = load_image('/Users/stamatiosorphanos/Documents/MCs_Thesis/SOD_Thesis/docs/Extended_Mask2Former/1.jpg')
+
+
+# Step 2: Initialise the Mask2Former Model 
+model = ExtendedMask2Former(num_classes=100, hidden_dim=256, num_queries=100, nheads=8, dim_feedforward=2048, dec_layers=1, mask_dim=256)
+
+# Step 3: Count the number of parameters
+total_params = sum(param.numel() for param in model.parameters())
+print('The total number of Extended Mask2Former parameters are: ', total_params)
+model.eval()  # Set the model to evaluation mode
+
+# Step 4: Pass the image through the model
+with torch.no_grad():
+    output = model(image)
+
+# Step 5: Visualize the feature maps
+print('\nThe predicted classes are the following:\n')
+print(output['pred_logits'])
 
-# # Step 6: Visualize the mask of the model
-# print('\nThe masks with shape: {} created are the following:\n'.format(output['pred_masks'].shape))
-# print(output['pred_masks'])
+# Step 6: Visualize the mask of the model
+print('\nThe masks with shape: {} created are the following:\n'.format(output['pred_masks'].shape))
+print(output['pred_masks'])
 
-# # Step 7: Visualize the bounding box of the model
-# print('\nThe bounding box with shape: {} created is the following:\n'.format(output['bounding_box'].shape))
-# print(output['bounding_box'])
+# Step 7: Visualize the bounding box of the model
+print('\nThe bounding box with shape: {} created is the following:\n'.format(output['bounding_box'].shape))
+print(output['bounding_box'])
 
-# # Step 8: Visualize the class scores
-# print('\nThe class scores with shape: {} created are the following:\n'.format(output['class_scores'].shape))
-# print(output['class_scores'])
+# Step 8: Visualize the class scores
+print('\nThe class scores with shape: {} created are the following:\n'.format(output['class_scores'].shape))
+print(output['class_scores'])
 
-# # # Step 9: Get the summary of the EFPN model
-# # print('\n\nThe summary of the EFPN model is the following: \n', model) 
-# # ------------------------------------------------------------------------------------------------------------------------------------
+# Step 9: Get the summary of the EFPN model
+print('\n\nThe summary of the EFPN model is the following: \n', model) 
+# ------------------------------------------------------------------------------------------------------------------------------------
 
 
 # Import data paths