Merge pull request #187 from Aydinhamedi/Alpha-b

Alpha b
Aydinhamedi · Mar 14, 2024 · bea0952 · bea0952
2 parents a73a30b + a9fa699
commit bea0952
Show file tree

Hide file tree

Showing 19 changed files with 175 additions and 210 deletions.
diff --git a/Archive/keras_applications_mod/models/convnext.py b/Archive/keras_applications_mod/models/convnext.py
@@ -228,12 +228,10 @@ def call(self, x):
 
     def get_config(self):
         config = super().get_config()
-        config.update(
-            {
-                "init_values": self.init_values,
-                "projection_dim": self.projection_dim,
-            }
-        )
+        config.update({
+            "init_values": self.init_values,
+            "projection_dim": self.projection_dim,
+        })
         return config
 
 

diff --git a/Archive/keras_applications_mod/models/imagenet_utils_test.py b/Archive/keras_applications_mod/models/imagenet_utils_test.py
@@ -74,13 +74,11 @@ def test_preprocess_input(self):
         self.assertAllClose(x, x2[..., ::-1])
         self.assertNotEqual(xint.astype("float").max(), xint2.max())
 
-    @parameterized.named_parameters(
-        [
-            {"testcase_name": "mode_torch", "mode": "torch"},
-            {"testcase_name": "mode_tf", "mode": "tf"},
-            {"testcase_name": "mode_caffe", "mode": "caffe"},
-        ]
-    )
+    @parameterized.named_parameters([
+        {"testcase_name": "mode_torch", "mode": "torch"},
+        {"testcase_name": "mode_tf", "mode": "tf"},
+        {"testcase_name": "mode_caffe", "mode": "caffe"},
+    ])
     def test_preprocess_input_symbolic(self, mode):
         # Test image batch
         x = np.random.uniform(0, 255, (2, 10, 10, 3))
@@ -131,13 +129,11 @@ def test_preprocess_input_symbolic(self, mode):
         out2 = model2.predict(x2[np.newaxis])[0]
         self.assertAllClose(out1, out2.transpose(1, 2, 0))
 
-    @parameterized.named_parameters(
-        [
-            {"testcase_name": "mode_torch", "mode": "torch"},
-            {"testcase_name": "mode_tf", "mode": "tf"},
-            {"testcase_name": "mode_caffe", "mode": "caffe"},
-        ]
-    )
+    @parameterized.named_parameters([
+        {"testcase_name": "mode_torch", "mode": "torch"},
+        {"testcase_name": "mode_tf", "mode": "tf"},
+        {"testcase_name": "mode_caffe", "mode": "caffe"},
+    ])
     def test_preprocess_input_symbolic_mixed_precision(self, mode):
         if not tf.__internal__.tf2.enabled():
             self.skipTest("The global policy can only be tested in TensorFlow 2")
@@ -152,18 +148,16 @@ def test_preprocess_input_symbolic_mixed_precision(self, mode):
         finally:
             set_global_policy("float32")
 
-    @parameterized.named_parameters(
-        [
-            {
-                "testcase_name": "channels_last_format",
-                "data_format": "channels_last",
-            },
-            {
-                "testcase_name": "channels_first_format",
-                "data_format": "channels_first",
-            },
-        ]
-    )
+    @parameterized.named_parameters([
+        {
+            "testcase_name": "channels_last_format",
+            "data_format": "channels_last",
+        },
+        {
+            "testcase_name": "channels_first_format",
+            "data_format": "channels_first",
+        },
+    ])
     def test_obtain_input_shape(self, data_format):
         # input_shape and default_size are not identical.
         with self.assertRaises(ValueError):

diff --git a/BETA_E_Model_T&T.ipynb b/BETA_E_Model_T&T.ipynb
@@ -622,7 +622,7 @@
     "            #  ┌-----------------┬--┬-------------------┐\n",
     "            #  │ 𝑦=2/(𝑥+1)^(𝑥+𝑉) ├OR┤ 𝑦=2/(𝑥+1)^(𝑥+1.4) │\n",
     "            #  └-----------------┴--┴-------------------┘\n",
-    "        print(f\">   Generating ADB[{i+1}/{ADBD}]...\")\n",
+    "        print(f\">   Generating ADB[{i + 1}/{ADBD}]...\")\n",
     "        # prepare an iterators to scale images\n",
     "        train_iterator = train_datagen.flow(x_train, y_train, batch_size=len(x_train))\n",
     "\n",
@@ -3214,12 +3214,10 @@
     "        layer.trainable = True\n",
     "\n",
     "    # Combine the output of the two base models\n",
-    "    combined = concatenate(\n",
-    "        [\n",
-    "            Dense(512, activation=\"relu\", kernel_regularizer=l2(0.01))(GlobalAveragePooling2D()(base_model1_out)),\n",
-    "            Dense(512, activation=\"relu\", kernel_regularizer=l2(0.01))(GlobalAveragePooling2D()(base_model2_out)),\n",
-    "        ]\n",
-    "    )\n",
+    "    combined = concatenate([\n",
+    "        Dense(512, activation=\"relu\", kernel_regularizer=l2(0.01))(GlobalAveragePooling2D()(base_model1_out)),\n",
+    "        Dense(512, activation=\"relu\", kernel_regularizer=l2(0.01))(GlobalAveragePooling2D()(base_model2_out)),\n",
+    "    ])\n",
     "\n",
     "    # adding CDL\n",
     "    Dense_L1 = Dense(1024, activation=\"relu\", kernel_regularizer=l2(0.02))(combined)\n",
@@ -9958,12 +9956,10 @@
     "\n",
     "        old_weights, old_biases = layer.get_weights()\n",
     "\n",
-    "        layer.set_weights(\n",
-    "            [\n",
-    "                weight_initializer(shape=old_weights.shape),\n",
-    "                bias_initializer(shape=len(old_biases)),\n",
-    "            ]\n",
-    "        )"
+    "        layer.set_weights([\n",
+    "            weight_initializer(shape=old_weights.shape),\n",
+    "            bias_initializer(shape=len(old_biases)),\n",
+    "        ])"
    ]
   },
   {
@@ -11032,18 +11028,18 @@
     "        optimizer = self.model.optimizer\n",
     "        if hasattr(optimizer, \"lr\"):\n",
     "            lr = tf.keras.backend.get_value(optimizer.lr)\n",
-    "            print(f\"\\nLearning rate for epoch {epoch+1} is {lr}\")\n",
+    "            print(f\"\\nLearning rate for epoch {epoch + 1} is {lr}\")\n",
     "        if hasattr(optimizer, \"momentum\"):\n",
     "            momentum = tf.keras.backend.get_value(optimizer.momentum)\n",
-    "            print(f\"Momentum for epoch {epoch+1} is {momentum}\")\n",
+    "            print(f\"Momentum for epoch {epoch + 1} is {momentum}\")\n",
     "        if logs:\n",
     "            val_loss = logs.get(\"val_loss\")\n",
     "            val_acc = logs.get(\"val_accuracy\")\n",
-    "            print(f\"Validation loss for epoch {epoch+1} is {val_loss}\")\n",
-    "            print(f\"Validation accuracy for epoch {epoch+1} is {val_acc}\")\n",
+    "            print(f\"Validation loss for epoch {epoch + 1} is {val_loss}\")\n",
+    "            print(f\"Validation accuracy for epoch {epoch + 1} is {val_acc}\")\n",
     "\n",
     "        print_Color_V2(\n",
-    "            f\"`red`<!--------------------------------------|Epoch`yellow` [{epoch+1}]`red` End|--------------------------------------!> `green`PBE↓\",\n",
+    "            f\"`red`<!--------------------------------------|Epoch`yellow` [{epoch + 1}]`red` End|--------------------------------------!> `green`PBE↓\",\n",
     "            start_char=\"`\",\n",
     "            end_char=\"`\",\n",
     "        )\n",
@@ -11538,7 +11534,7 @@
     "    plt.xlabel(\"Epoch (within chunk)\")\n",
     "    plt.ylabel(\"Average Value\")\n",
     "    plt.title(\"Average Validation Accuracy and Loss for Each Epoch Across Chunks (Ignoring First Chunk)\")\n",
-    "    plt.xticks(epoch_indices, [f\"Epoch {i+1}\" for i in epoch_indices])  # Set x-tick labels to epoch numbers\n",
+    "    plt.xticks(epoch_indices, [f\"Epoch {i + 1}\" for i in epoch_indices])  # Set x-tick labels to epoch numbers\n",
     "    plt.legend()\n",
     "\n",
     "    plt.tight_layout()\n",

diff --git a/Exports/V5/Python_EPO.py b/Exports/V5/Python_EPO.py
@@ -595,18 +595,14 @@ def Debug_img_Save(img, id="DEF"):
 mean_intensity_mean = reshaped_data_mean.mean(axis=-1)
 
 # Stack the mean intensity with the reshaped data
-data_with_mean_pixels = np.hstack(
-    [
-        reshaped_data_pixels,
-        mean_intensity_pixels.reshape(-1, 1),
-    ]
-)
-data_with_mean_mean = np.hstack(
-    [
-        reshaped_data_mean,
-        mean_intensity_mean.reshape(-1, 1),
-    ]
-)
+data_with_mean_pixels = np.hstack([
+    reshaped_data_pixels,
+    mean_intensity_pixels.reshape(-1, 1),
+])
+data_with_mean_mean = np.hstack([
+    reshaped_data_mean,
+    mean_intensity_mean.reshape(-1, 1),
+])
 
 # Calculate Z-scores
 z_scores_pixels = np.abs(zscore(data_with_mean_pixels, axis=0))
@@ -855,12 +851,10 @@ def Combo_Model(freeze_layers1, freeze_layers2):
         layer.trainable = True
 
     # Combine the output of the two base models
-    combined = concatenate(
-        [
-            GlobalAveragePooling2D()(base_model1_out),
-            GlobalAveragePooling2D()(base_model2_out),
-        ]
-    )
+    combined = concatenate([
+        GlobalAveragePooling2D()(base_model1_out),
+        GlobalAveragePooling2D()(base_model2_out),
+    ])
 
     # adding CDL
     Dense_L1 = Dense(1024, activation="relu", kernel_regularizer=l2(0.03))(combined)
@@ -1153,12 +1147,10 @@ def Eff_B7_NS():
 
         old_weights, old_biases = layer.get_weights()
 
-        layer.set_weights(
-            [
-                weight_initializer(shape=old_weights.shape),
-                bias_initializer(shape=len(old_biases)),
-            ]
-        )
+        layer.set_weights([
+            weight_initializer(shape=old_weights.shape),
+            bias_initializer(shape=len(old_biases)),
+        ])
 
 
 # %% [markdown]

diff --git a/Exports/V7/Python_EPO.py b/Exports/V7/Python_EPO.py
@@ -592,18 +592,14 @@ def Debug_img_Save(img, id="DEF"):
 mean_intensity_mean = reshaped_data_mean.mean(axis=-1)
 
 # Stack the mean intensity with the reshaped data
-data_with_mean_pixels = np.hstack(
-    [
-        reshaped_data_pixels,
-        mean_intensity_pixels.reshape(-1, 1),
-    ]
-)
-data_with_mean_mean = np.hstack(
-    [
-        reshaped_data_mean,
-        mean_intensity_mean.reshape(-1, 1),
-    ]
-)
+data_with_mean_pixels = np.hstack([
+    reshaped_data_pixels,
+    mean_intensity_pixels.reshape(-1, 1),
+])
+data_with_mean_mean = np.hstack([
+    reshaped_data_mean,
+    mean_intensity_mean.reshape(-1, 1),
+])
 
 # Calculate Z-scores
 z_scores_pixels = np.abs(zscore(data_with_mean_pixels, axis=0))
@@ -860,12 +856,10 @@ def Combo_Model(freeze_layers1, freeze_layers2):
         layer.trainable = True
 
     # Combine the output of the two base models
-    combined = concatenate(
-        [
-            Dense(512, activation="relu", kernel_regularizer=l2(0.02))(GlobalAveragePooling2D()(base_model1_out)),
-            Dense(512, activation="relu", kernel_regularizer=l2(0.02))(GlobalAveragePooling2D()(base_model2_out)),
-        ]
-    )
+    combined = concatenate([
+        Dense(512, activation="relu", kernel_regularizer=l2(0.02))(GlobalAveragePooling2D()(base_model1_out)),
+        Dense(512, activation="relu", kernel_regularizer=l2(0.02))(GlobalAveragePooling2D()(base_model2_out)),
+    ])
 
     # adding CDL
     Dense_L1 = Dense(1024, activation="relu", kernel_regularizer=l2(0.03))(combined)
@@ -1411,12 +1405,10 @@ def load_model(filename):
 
         old_weights, old_biases = layer.get_weights()
 
-        layer.set_weights(
-            [
-                weight_initializer(shape=old_weights.shape),
-                bias_initializer(shape=len(old_biases)),
-            ]
-        )
+        layer.set_weights([
+            weight_initializer(shape=old_weights.shape),
+            bias_initializer(shape=len(old_biases)),
+        ])
 
 
 # %% [markdown]

diff --git a/Interface/GUI/Data/GUI_main.py b/Interface/GUI/Data/GUI_main.py
@@ -29,7 +29,7 @@
 sg.popup_auto_close("Loading GUI...", non_blocking=True, auto_close_duration=2, no_titlebar=False)
 # pylib
 try:
-    import re
+    # import re # noqa: F401
     import time
     import cv2
     import json
@@ -38,14 +38,17 @@
     import hashlib
     import pydicom
     import cpuinfo
-    import difflib
-    import inspect
+
+    # import difflib # noqa: F401
+    # import inspect # noqa: F401
     import traceback
     import subprocess
     import threading
     import requests
+    import numpy as np
     from tqdm import tqdm
-    from time import sleep
+
+    # from time import sleep # noqa: F401
     from loguru import logger
     from tkinter import filedialog
     from datetime import datetime
@@ -57,10 +60,10 @@
     os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"
     # Utils
     from Utils.Grad_cam import make_gradcam_heatmap
-    from Utils.print_color_V2_NEW import print_Color_V2
+    from Utils.print_color_V2_NEW import print_Color_V2  # noqa: F401
     from Utils.print_color_V1_OLD import print_Color
     from Utils.FixedDropout import FixedDropout
-    from Utils.Other import *
+    from Utils.Other import *  # noqa: F403
 except (ImportError, NameError):
     print("Failed to load the GUI libs")
     print("detailed error message:")
@@ -134,9 +137,8 @@ def is_updated(self):
 GUI_layout_Tab_main = [
     [sg.Text("Enter the image dir:", font=(None, 10, "bold"))],
     [
-        sg.Input(key="-INPUT_IMG_dir-"),
+        sg.Input(key="-INPUT_IMG_dir-", enable_events=True, size=(48, 1)),
         sg.Button("Browse", key="-BUTTON_BROWSE_IMG_dir-"),
-        sg.Button("Ok", key="-BUTTON_OK_IMG_dir-"),
     ],
     [sg.Text("Log:", font=(None, 10, "bold"))],
     [sg.Multiline(key="-OUTPUT_ST-", size=(54, 6), autoscroll=True)],
@@ -538,7 +540,8 @@ def CI_liid(img_dir, Show_DICOM_INFO: bool = True) -> str:
     try:
         _, file_extension = os.path.splitext(img_dir)
     except TypeError:
-        file_extension = "TEMP FILE EXTENSION"
+        logger.warning("CI_liid>>ERROR: Invalid file format. Please provide an image file. (Extension Extractiion Failed)")
+        return "ERROR: Invalid file format. Please provide an image file. (Extension Extractiion Failed)"
     if file_extension.upper()[1:] not in IMG_AF:
         logger.warning("CI_liid>>ERROR: Invalid file format. Please provide an image file.")
         return "ERROR: Invalid file format. Please provide an image file."
@@ -567,7 +570,7 @@ def CI_liid(img_dir, Show_DICOM_INFO: bool = True) -> str:
                     GUI_layout_DICOM_Info_Window.finalize()
             else:
                 img = Image.open(img_dir).resize((IMG_RES[1], IMG_RES[0]))
-        except NameError:
+        except (NameError, FileNotFoundError):
             logger.warning("CI_liid>>ERROR: Invalid file dir. Please provide an image file.")
             return "ERROR: Invalid file dir. Please provide an image file."
         else:
@@ -777,7 +780,7 @@ def main() -> None:
             GUI_window["-INPUT_IMG_dir-"].update(IMG_dir)
 
         # Handle event for confirming the selected image directory
-        if event == "-BUTTON_OK_IMG_dir-":
+        if event == "-INPUT_IMG_dir-":
             # Retrieve the image directory from the input field and update the display
             IMG_dir = GUI_window["-INPUT_IMG_dir-"].get()
             GUI_window["-INPUT_IMG_dir-"].update(IMG_dir)