Upload 4 files

100-epoch with regularization.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec769e8790691bbf2f5445b3a338095db7af87939df0f323eb01b91234daf5b3
+size 2554400

cnn_train.py

@@ -0,0 +1,66 @@
+import os
+os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
+
+import warnings
+warnings.filterwarnings('ignore')
+
+import tensorflow as tf
+from keras.models import Sequential
+from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout
+from keras.callbacks import EarlyStopping
+import numpy as np
+
+np.random.seed(1337)
+classifier = Sequential()
+
+classifier.add(Conv2D(32, (3, 3), input_shape=(128, 128, 3), activation='relu'))
+classifier.add(MaxPooling2D(pool_size=(2, 2)))
+classifier.add(Conv2D(16, (3, 3), activation='relu'))
+classifier.add(MaxPooling2D(pool_size=(2, 2)))
+classifier.add(Conv2D(8, (3, 3), activation='relu'))
+classifier.add(MaxPooling2D(pool_size=(2, 2)))
+classifier.add(Flatten())
+classifier.add(Dense(128, activation='relu'))
+classifier.add(Dropout(0.5))
+classifier.add(Dense(10, activation='softmax'))
+
+classifier.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
+print(classifier.summary())
+
+train_dir = '/home/vignesh/tomato_data/train'
+val_dir = '/home/vignesh/tomato_data/val'
+
+train_data_raw = tf.keras.utils.image_dataset_from_directory(
+    train_dir,
+    labels='inferred',
+    label_mode='categorical',
+    image_size=(128, 128),
+    batch_size=32,
+    shuffle=True
+)
+class_names = train_data_raw.class_names  # Get class names before mapping
+train_data = train_data_raw.map(lambda x, y: (x / 255.0, y)).prefetch(tf.data.AUTOTUNE)
+
+val_data = tf.keras.utils.image_dataset_from_directory(
+    val_dir,
+    labels='inferred',
+    label_mode='categorical',
+    image_size=(128, 128),
+    batch_size=32,
+    shuffle=False
+)
+val_data = val_data.map(lambda x, y: (x / 255.0, y)).prefetch(tf.data.AUTOTUNE)
+
+print({name: idx for idx, name in enumerate(class_names)})
+
+# Early stopping callback
+early_stop = EarlyStopping(monitor='val_loss', patience=10,restore_best_weights=True)
+
+classifier.fit(
+    train_data,
+    epochs=30,
+    validation_data=val_data,
+)
+
+classifier.save('keras_potato_trained_model(2.h5')
+print('Saved trained model as %s ' % 'keras_potato_trained_model.h5')

requirements.txt

@@ -1,3 +1,5 @@
-altair
-pandas
-streamlit
+streamlit>=1.20.0
+numpy>=1.23.0
+pillow>=9.0.0
+tensorflow>=2.10.0,<2.16.0
+h5py>=3.7.0 

streamlit.py

@@ -0,0 +1,38 @@
+import streamlit as st
+import tensorflow as tf
+import numpy as np
+from PIL import Image
+
+st.title('🍅 Simple Tomato Leaf Disease Classifier')
+
+@st.cache_resource
+def load_model():
+    return tf.keras.models.load_model('100-epoch with regularization.h5')
+
+model = load_model()
+
+# Class names (update if your classes are different)
+class_names = [
+    'Tomato___Bacterial_spot',
+    'Tomato___Early_blight',
+    'Tomato___Late_blight',
+    'Tomato___Leaf_Mold',
+    'Tomato___Septoria_leaf_spot',
+    'Tomato___Spider_mites Two-spotted_spider_mite',
+    'Tomato___Target_Spot',
+    'Tomato___Tomato_Yellow_Leaf_Curl_Virus',
+    'Tomato___Tomato_mosaic_virus',
+    'Tomato___healthy'
+]
+
+uploaded_file = st.file_uploader('Upload a tomato leaf image', type=['jpg', 'jpeg', 'png'])
+
+if uploaded_file is not None:
+    image = Image.open(uploaded_file).convert('RGB')
+    st.image(image, caption='Uploaded Image', use_column_width=True)
+    img = image.resize((128, 128))
+    img_array = np.array(img) / 255.0
+    img_array = np.expand_dims(img_array, axis=0)
+    preds = model.predict(img_array)
+    pred_class = np.argmax(preds, axis=1)[0]
+    st.success(f'Predicted Class: {class_names[pred_class]}')