Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Support
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
22_23-J 65
Project overview
Project overview
Details
Activity
Releases
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Issues
0
Issues
0
List
Boards
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Analytics
Analytics
CI / CD
Repository
Value Stream
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
22_23-J 65
22_23-J 65
Commits
b00dfb88
Commit
b00dfb88
authored
Jan 28, 2023
by
dulanthaM
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
data visualization is completed
parent
975e121e
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
375 additions
and
6 deletions
+375
-6
Backend/SF_Dulantha_Dificiency_identificationV1.ipynb
Backend/SF_Dulantha_Dificiency_identificationV1.ipynb
+375
-6
No files found.
Backend/SF_Dulantha_Dificiency_identificationV1.ipynb
View file @
b00dfb88
...
@@ -19,7 +19,15 @@
...
@@ -19,7 +19,15 @@
"id": "lDXyIngTMaXf",
"id": "lDXyIngTMaXf",
"outputId": "eb920137-7143-4c2c-e43e-a0aa82d93b49"
"outputId": "eb920137-7143-4c2c-e43e-a0aa82d93b49"
},
},
"outputs": [],
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Mounted at /content/drive\n"
]
}
],
"source": [
"source": [
"from google.colab import drive\n",
"from google.colab import drive\n",
"drive.mount('/content/drive')"
"drive.mount('/content/drive')"
...
@@ -46,7 +54,20 @@
...
@@ -46,7 +54,20 @@
"id": "widI8pNLhUci",
"id": "widI8pNLhUci",
"outputId": "13c35bb6-dd84-4ffd-f585-402d31aeaa41"
"outputId": "13c35bb6-dd84-4ffd-f585-402d31aeaa41"
},
},
"outputs": [],
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
"Collecting patool\n",
" Downloading patool-1.12-py2.py3-none-any.whl (77 kB)\n",
"\u001b[K |████████████████████████████████| 77 kB 5.7 MB/s \n",
"\u001b[?25hInstalling collected packages: patool\n",
"Successfully installed patool-1.12\n"
]
}
],
"source": [
"source": [
"!pip install patool"
"!pip install patool"
]
]
...
@@ -62,7 +83,31 @@
...
@@ -62,7 +83,31 @@
"id": "U-MmiQnjF9gG",
"id": "U-MmiQnjF9gG",
"outputId": "3d634674-631e-4ed7-d8d7-708c1f9050f5"
"outputId": "3d634674-631e-4ed7-d8d7-708c1f9050f5"
},
},
"outputs": [],
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"patool: Extracting /content/drive/MyDrive/RP_SmartFarmer/Dulantha Madhushan - Nutrient deficiency/Dataset/Dificiency Data.rar ...\n",
"patool: running /usr/bin/unrar x -- \"/content/drive/MyDrive/RP_SmartFarmer/Dulantha Madhushan - Nutrient deficiency/Dataset/Dificiency Data.rar\"\n",
"patool: with cwd='/content'\n",
"patool: ... /content/drive/MyDrive/RP_SmartFarmer/Dulantha Madhushan - Nutrient deficiency/Dataset/Dificiency Data.rar extracted to `/content'.\n"
]
},
{
"data": {
"application/vnd.google.colaboratory.intrinsic+json": {
"type": "string"
},
"text/plain": [
"'/content'"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"source": [
"import patoolib\n",
"import patoolib\n",
"patoolib.extract_archive(data_location, outdir=\"/content\")"
"patoolib.extract_archive(data_location, outdir=\"/content\")"
...
@@ -78,7 +123,23 @@
...
@@ -78,7 +123,23 @@
"id": "_YdAEKT6GSoX",
"id": "_YdAEKT6GSoX",
"outputId": "4fdd684c-a8b2-4fa3-ee3e-91cd0c9ec280"
"outputId": "4fdd684c-a8b2-4fa3-ee3e-91cd0c9ec280"
},
},
"outputs": [],
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"There are 2 directories and 0 images in 'Dificiency Data'.\n",
"There are 3 directories and 0 images in 'Dificiency Data/Trainig'.\n",
"There are 0 directories and 280 images in 'Dificiency Data/Trainig/potassium deficiency'.\n",
"There are 0 directories and 104 images in 'Dificiency Data/Trainig/phosphorus dificiency'.\n",
"There are 0 directories and 196 images in 'Dificiency Data/Trainig/magnesium deficiency'.\n",
"There are 3 directories and 0 images in 'Dificiency Data/Testing'.\n",
"There are 0 directories and 112 images in 'Dificiency Data/Testing/potassium deficiency'.\n",
"There are 0 directories and 104 images in 'Dificiency Data/Testing/phosphorus dificiency'.\n",
"There are 0 directories and 157 images in 'Dificiency Data/Testing/magnesium deficiency'.\n"
]
}
],
"source": [
"source": [
"import os\n",
"import os\n",
"for dirpath , dirnames , filenames in os.walk(\"Dificiency Data\"):\n",
"for dirpath , dirnames , filenames in os.walk(\"Dificiency Data\"):\n",
...
@@ -95,7 +156,15 @@
...
@@ -95,7 +156,15 @@
"id": "gLo3FV6OGVP0",
"id": "gLo3FV6OGVP0",
"outputId": "c9f815fd-004f-44d5-dba0-745f657c09fb"
"outputId": "c9f815fd-004f-44d5-dba0-745f657c09fb"
},
},
"outputs": [],
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"['magnesium deficiency' 'phosphorus dificiency' 'potassium deficiency']\n"
]
}
],
"source": [
"source": [
"train_dir = \"Dificiency Data/Trainig\"\n",
"train_dir = \"Dificiency Data/Trainig\"\n",
"test_dir = \"Dificiency Data/Testing\"\n",
"test_dir = \"Dificiency Data/Testing\"\n",
...
@@ -109,6 +178,15 @@
...
@@ -109,6 +178,15 @@
"print(class_names)"
"print(class_names)"
]
]
},
},
{
"cell_type": "markdown",
"metadata": {
"id": "HvDEsB_fj7_7"
},
"source": [
"# 🟨 Data Visualization"
]
},
{
{
"cell_type": "code",
"cell_type": "code",
"execution_count": null,
"execution_count": null,
...
@@ -146,12 +224,303 @@
...
@@ -146,12 +224,303 @@
"id": "tVx10O9CGXn2",
"id": "tVx10O9CGXn2",
"outputId": "173a0a39-9cfa-4f76-cf91-9e67e7da54d0"
"outputId": "173a0a39-9cfa-4f76-cf91-9e67e7da54d0"
},
},
"outputs": [],
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Image Shape:(255, 255, 3)\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"source": [
"import random\n",
"import random\n",
"img = view_random_image(target_dir=train_dir , \n",
"img = view_random_image(target_dir=train_dir , \n",
" target_class = random.choice(class_names))"
" target_class = random.choice(class_names))"
]
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "8uTOAKV80c_S"
},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "oIC5RKBhG-t-",
"outputId": "2e12d877-3d10-4f40-9706-a593a1da7494"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 580 images belonging to 3 classes.\n",
"Found 373 images belonging to 3 classes.\n"
]
}
],
"source": [
"from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
"\n",
"#Creating Image Data Generators for Training Data with augmentation\n",
"train_data_gen = ImageDataGenerator(rescale=1/255.,\n",
" #preprocessing_function=to_grayscale_then_rgb,\n",
" rotation_range = 0.25,\n",
" shear_range = 0.25,\n",
" zoom_range = 0.25,\n",
" width_shift_range=0.3,\n",
" height_shift_range = 0.3,\n",
" horizontal_flip= True)\n",
"\n",
"#Create ImageDatagenerator for testing data\n",
"test_data_gen = ImageDataGenerator(rescale=1/255.)\n",
"\n",
"#Import and Transform/pre process the data\n",
"train_data_multi = train_data_gen.flow_from_directory(train_dir,\n",
" target_size = (224,224),\n",
" batch_size = 32,\n",
" class_mode = 'categorical',\n",
" shuffle = True)\n",
"test_data_multi = test_data_gen.flow_from_directory(test_dir,\n",
" target_size = (224, 224),\n",
" batch_size = 32,\n",
" class_mode = 'categorical', #This will gives us one-hot encoded data\n",
" shuffle = True)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "z_3CSlVYgcU7"
},
"source": [
"#TransferLoearning Model"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "UMfVmghAgha4"
},
"outputs": [],
"source": [
"efficientnet_url = \"https://tfhub.dev/tensorflow/efficientnet/b0/feature-vector/1\""
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "ypHVBoUagk76"
},
"outputs": [],
"source": [
"import tensorflow as tf\n",
"import tensorflow_hub as hub\n",
"from tensorflow.keras import layers"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "cO9abOkFgnNs"
},
"outputs": [],
"source": [
"def create_model(model_url , num_classes=7):\n",
"\n",
" #Download the pretrained model and save it as a keras layer\n",
" feature_extractor_layer = hub.KerasLayer(model_url,\n",
" trainable = False, #freeze the already learned patterns \n",
" name = \"feature_extraction_layer\",\n",
" input_shape = (224, 224,3)) \n",
" #Create our own model\n",
" model = tf.keras.Sequential([\n",
" feature_extractor_layer,\n",
" layers.Dense(num_classes , activation=\"softmax\" , name=\"output_layer\")\n",
" ])\n",
"\n",
" return model"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "rX3XVcMpgoPE"
},
"outputs": [],
"source": [
"#Create Resnet Model\n",
"efficientNet_model = create_model(efficientnet_url , \n",
" num_classes = 3)\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "HzB5iBCig11l",
"outputId": "8fa480b4-4cce-46e5-9abb-98e6fd482b77"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/5\n",
"19/19 [==============================] - 38s 1s/step - loss: 0.8127 - accuracy: 0.6103 - val_loss: 0.5547 - val_accuracy: 0.8231\n",
"Epoch 2/5\n",
"19/19 [==============================] - 22s 1s/step - loss: 0.4954 - accuracy: 0.8190 - val_loss: 0.3970 - val_accuracy: 0.8794\n",
"Epoch 3/5\n",
"19/19 [==============================] - 21s 1s/step - loss: 0.4086 - accuracy: 0.8362 - val_loss: 0.3707 - val_accuracy: 0.8767\n",
"Epoch 4/5\n",
"19/19 [==============================] - 21s 1s/step - loss: 0.3657 - accuracy: 0.8621 - val_loss: 0.3174 - val_accuracy: 0.8874\n",
"Epoch 5/5\n",
"19/19 [==============================] - 21s 1s/step - loss: 0.3289 - accuracy: 0.8810 - val_loss: 0.2937 - val_accuracy: 0.9115\n"
]
}
],
"source": [
"#Compile our resnet model\n",
"efficientNet_model.compile(loss='categorical_crossentropy',\n",
" optimizer = tf.keras.optimizers.Adam(),\n",
" metrics=[\"accuracy\"])\n",
"#Fitting the model\n",
"resnet_hist = efficientNet_model.fit(train_data_multi,\n",
" epochs=5,\n",
" steps_per_epoch=len(train_data_multi),\n",
" validation_data = test_data_multi,\n",
" validation_steps = len(test_data_multi)\n",
" )"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "eTHdQekBuSoD"
},
"outputs": [],
"source": [
"efficientNet_model.save('Dificiency_Identification_v1.h5')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "Yjc9mF38uSfK"
},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "gx3orjVQMLLt"
},
"source": [
"#Make Predictions"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "0AiHHV3BhBjK"
},
"outputs": [],
"source": [
"#Create a function to improt an image and resize it to be able to used with our model\n",
"def load_and_prep_image(filename , img_shape=224):\n",
" \"\"\"\n",
" Reads an image from file name , turns it into a tensor and reshapes it to (img_shape , color_channel\n",
" \"\"\"\n",
" print('FILe Name type - ', type(filename))\n",
" #Read in the image\n",
" img = tf.io.read_file(filename)\n",
" print('img 1- ', type(img))\n",
"\n",
" #Decode the read file into a tensor\n",
" img = tf.image.decode_image(img)\n",
" print('img 2- ', type(img))\n",
"\n",
" #Resiz the Image\n",
" img = tf.image.resize(img , size=[img_shape , img_shape])\n",
"\n",
" #Re scale the image - get all values between 0 and 255\n",
" img = img/255.\n",
"\n",
" return img\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "NJWZDQ58hV_T"
},
"outputs": [],
"source": [
"def pred_and_plot(model, filename , class_names = class_names):\n",
" \"\"\"\n",
" Import an image located at filename , makes a prediction with model,\n",
" and plot the image with the prediced class as the title\n",
" \"\"\"\n",
" #Import the target image and pre-process\n",
" img = load_and_prep_image(filename)\n",
" print('FILe Name type - ', type(filename))\n",
"\n",
" #Make a Prediction\n",
" pred = model.predict(tf.expand_dims(img , axis=0))\n",
"\n",
" #Get the predicted class\n",
" pred_class = class_names[np.argmax(tf.round(pred))]\n",
"\n",
" #Plot the image and the predicted class\n",
" plt.imshow(img)\n",
" plt.title(f\"Prediction :{pred_class}\")\n",
" plt.axis(False)"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"id": "rZBWANZuAUPg"
},
"outputs": [],
"source": [
"# image_path5 = \"\"\n",
"# pred_and_plot(resnet_model , image_path5)"
]
}
}
],
],
"metadata": {
"metadata": {
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment