initial commit

find2.py

@@ -0,0 +1,54 @@
+#!/usr/bin/python3
+
+img_size = (150, 150)
+
+from tensorflow.keras import models as m
+from tensorflow.keras import layers as l
+from tensorflow.keras import optimizers as o
+from PIL import Image
+from sys import argv
+from os import listdir as ls
+import numpy as np
+
+model = m.Sequential([
+    l.Input(shape = (*img_size, 3)),
+
+    l.Conv2D(96, (11, 11), strides = 4, activation = "relu"),
+    l.BatchNormalization(),
+    l.MaxPooling2D((3, 3), strides = 2),
+
+    l.Conv2D(192, (5, 5), activation = "relu", padding = "same"),
+    l.BatchNormalization(),
+    l.MaxPooling2D((3, 3), strides = 2),
+
+    l.Conv2D(256, (3, 3), activation = "relu", padding = "same"),
+    l.Conv2D(256, (3, 3), activation = "relu", padding = "same"),
+    l.Conv2D(160, (3, 3), activation = "relu", padding = "same"),
+    l.BatchNormalization(),
+    l.MaxPooling2D((3, 3), strides = 2),
+
+    l.Flatten(),
+    l.Dense(1024, activation = "relu"),
+    l.Dropout(0.5),
+    l.Dense(1024, activation = "relu"),
+    l.Dropout(0.5),
+    l.Dense(10, activation = "softmax"),
+])
+
+model.compile(optimizer = o.Adam(learning_rate = 0.0001),
+              loss = "categorical_crossentropy",
+              metrics = ["accuracy"])
+
+model.load_weights("w2.weights.h5")
+
+if len(argv) >= 2:
+    for i in argv[1:]:
+        with Image.open(i) as im:
+            im = im.resize((150, 150), Image.Resampling.LANCZOS)
+            im = np.divide(np.array(im),
+                           np.array(255.))
+
+            res = model.predict(np.array([im]))
+            print(res)
+            print(np.argmax(res))
+            print(sorted(ls("../ds/raw-img"))[np.argmax(res)])

test1.py

@@ -0,0 +1,62 @@
+#!/usr/bin/python3
+
+from tensorflow.keras.preprocessing.image import ImageDataGenerator
+
+img_size = (227, 227)
+batch_size = 128
+extract_path="../ds/raw-img"
+
+datagen = ImageDataGenerator(
+        rescale=1.0/255,
+        validation_split=0.2
+)
+
+def __dg(subset):
+    return datagen.flow_from_directory(extract_path,
+                                       target_size = img_size,
+                                       batch_size = batch_size,
+                                       class_mode = "categorical",
+                                       subset = subset,
+                                       shuffle = True)
+
+train_generator = __dg("training")
+val_generator = __dg("validation")
+
+
+from tensorflow.keras import models as m
+from tensorflow.keras import layers as l
+from tensorflow.keras import optimizers as o
+
+model = m.Sequential([
+    l.Input(shape=(227, 227, 3)),
+    l.Conv2D(96, (11, 11), strides=4, activation='relu'),
+    l.BatchNormalization(),
+    l.MaxPooling2D((3, 3), strides=2),
+
+    l.Conv2D(256, (5, 5), activation='relu', padding='same'),
+    l.BatchNormalization(),
+    l.MaxPooling2D((3, 3), strides=2),
+    l.Conv2D(384, (3, 3), activation='relu', padding='same'),
+    l.Conv2D(384, (3, 3), activation='relu', padding='same'),
+    l.Conv2D(256, (3, 3), activation='relu', padding='same'),
+    l.BatchNormalization(),
+    l.MaxPooling2D((3, 3), strides=2),
+    l.Flatten(),
+    l.Dense(4096, activation='relu'),
+    l.Dropout(0.5),
+    l.Dense(4096, activation='relu'),
+    l.Dropout(0.5),
+    l.Dense(10, activation='softmax'),
+])
+
+model.compile(optimizer = o.Adam(learning_rate = 0.0001),
+              loss = 'categorical_crossentropy',
+              metrics = ['accuracy'])
+
+print(model.summary())
+
+model.fit(train_generator,
+          epochs = 3,
+          validation_data = val_generator)
+
+model.save_weights("w1.weights.h5")

test2.py

@@ -0,0 +1,64 @@
+#!/usr/bin/python3
+
+from tensorflow.keras.preprocessing.image import ImageDataGenerator
+
+img_size = (150, 150)
+batch_size = 128
+extract_path="../ds/raw-img"
+
+datagen = ImageDataGenerator(
+        rescale=1.0/255,
+        validation_split=0.2
+)
+
+def __dg(subset):
+    return datagen.flow_from_directory(extract_path,
+                                       target_size = img_size,
+                                       batch_size = batch_size,
+                                       class_mode = "categorical",
+                                       subset = subset,
+                                       shuffle = True)
+
+train_generator = __dg("training")
+val_generator = __dg("validation")
+
+
+from tensorflow.keras import models as m
+from tensorflow.keras import layers as l
+from tensorflow.keras import optimizers as o
+
+model = m.Sequential([
+    l.Input(shape=(150, 150, 3)),
+    l.Conv2D(96, (11, 11), strides=4, activation='relu'),
+    l.BatchNormalization(),
+    l.MaxPooling2D((3, 3), strides=2),
+
+    l.Conv2D(192, (5, 5), activation='relu', padding='same'),
+    l.BatchNormalization(),
+    l.MaxPooling2D((3, 3), strides=2),
+    l.Conv2D(256, (3, 3), activation='relu', padding='same'),
+    l.Conv2D(256, (3, 3), activation='relu', padding='same'),
+    l.Conv2D(160, (3, 3), activation='relu', padding='same'),
+    l.BatchNormalization(),
+    l.MaxPooling2D((3, 3), strides=2),
+    l.Flatten(),
+    l.Dense(1024, activation='relu'),
+    l.Dropout(0.5),
+    l.Dense(1024, activation='relu'),
+    l.Dropout(0.5),
+    l.Dense(10, activation='softmax'),
+])
+
+model.compile(optimizer = o.Adam(learning_rate = 0.0001),
+              loss = 'categorical_crossentropy',
+              metrics = ['accuracy'])
+
+print(model.summary())
+
+model.load_weights("w2.weights.h5")
+
+model.fit(train_generator,
+          epochs = 10,
+          validation_data = val_generator)
+
+model.save_weights("w2.weights.h5")