{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "QquaYpB00hfs" }, "source": [ "[](http://edenlibrary.ai/)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Weeds Identification-Transfer Learning-5" ] }, { "cell_type": "markdown", "metadata": { "id": "yp-OgP370hft" }, "source": [ "## Instructions\n", "To run any of Eden's notebooks, please check the guides on our [Wiki page](https://github.com/Eden-Library-AI/eden_library_notebooks/wiki).
\n", "There you will find instructions on how to deploy the notebooks on [your local system](https://github.com/Eden-Library-AI/eden_library_notebooks/wiki/Deploy-Notebooks-Locally), on [Google Colab](https://github.com/Eden-Library-AI/eden_library_notebooks/wiki/Deploy-Notebooks-on-GColab), or on [MyBinder](https://github.com/Eden-Library-AI/eden_library_notebooks/wiki/Deploy-Notebooks-on-MyBinder), as well as other useful links, troubleshooting tips, and more.
\n", "For this notebook you will need to download the **Cotton-100619-Healthy-zz-V1-20210225102300**, **Black nightsade-220519-Weed-zz-V1-20210225102034**, **Tomato-240519-Healthy-zz-V1-20210225103740** and **Velvet leaf-220519-Weed-zz-V1-20210225104123** datasets from [Eden Library](https://edenlibrary.ai/datasets), and you may want to use the **eden_tensorflow_transfer_learning.yml** file to recreate a suitable conda environment.\n", "\n", "**Note:** If you find any issues while executing the notebook, don't hesitate to open an issue on Github. We will try to reply as soon as possible." ] }, { "cell_type": "markdown", "metadata": { "id": "dwqUKc-L0hft" }, "source": [ "## Background\n", "\n", "In this notebook, we are going to cover a technique called **Transfer Learning**, which generally refers to a process where a machine learning model is trained on one problem, and afterwards, it is reused in some way on a second (possibly) related problem (Bengio, 2012). Specifically, in **deep learning**, this technique is used by training only some layers of the pre-trained network. Its promise is that the training will be more efficient and in the best of the cases the performance will be better compared to a model trained from scratch. In this example we are using MobileNet architectures.\n", "\n", "**MobileNet V1**, with its appearance in 2017, started a new path of deep learning research in computer vision: models that can run in **embedded devices**. This lead to several important works including ShuffleNet(V1 and V2), MNasNet, CondenseNet, and EffNet (among others). Due to its success, new versions of MobileNet have also appeared more recently. These new versions included new advances, such as **Inverted Residual Block** or the use of **Neural Architecture Search** for optimizing the creation of new neural components. **MobileNet Versions 2 (Sandler et al., 2018) and 3 (Howard et al., 2019)** are used as an example in this notebook.\n", "\n", "In **agriculture**, since **weeds** compete with crops in the domain of space, light and nutrients, they are an important problem that can lead to a poorer harvest by farmers. To avoid this, weeds should be removed at every step of the growth, but especially at the initial stages. For that reason, identifying weeds accurately by deep learning has arisen as an important objective. Related to this, we can find the disease detection problem, where transfer learning has also been used. Among the most relevant recent works, we can find:\n", "\n", "**Wang et al., (2017)** used transfer learning in order to obtain the best neural-based method for disease detection in plants. They extended the apple black rot images in the PlantVillage dataset, which were further annotated by botanists with four severity stages as ground truth. Then, they evaluated the performances of shallow networks trained from scratch and deep models fine-tuned by transfer learning. Their best model was the **VGG16** architecture trained with transfer learning, which yielded an overall accuracy of 90.4% on the hold-out test set. In **Mehdipour-Ghazi et al., (2017)**, the authors used the plant datasets of LifeCLEF 2015. Three popular deep learning architectures were evaluated: **GoogLeNet, AlexNet, and VGGNet**. Their best combined system (a combination of GoogleNet and VGGNet) achieved an overall accuracy of 80% on the validation set and an overall inverse rank score of 0.752 on the official test set. In **Suh et al., (2018)**, the authors compared different transfer learning approaches in order to find a suitable approach for weed detection (volunteer potato). Their highest classification accuracy for **AlexNet** was 98.0%. Comparing different networks, their highest classification accuracy was 98.7%, which was obtained with **VGG-19**. Additionally, all scenarios and pre-trained networks were feasible for real-time applications (classification time < 0.1 s). Another relevant study has been performed by **Kounalakis et al., (2019)** where they evaluated transfer learning by a combination of CNN-based feature extraction and linear classifiers to recognize rumex under real-world conditions. Their best system (**Inception_v1**+L2regLogReg) achieved an accuracy of 96.13 with a false positive rate of 3.62. In **Too et al., (2019)**, the authors used transfer learning achieving a performance of 99.75% with the **DenseNet** architecture. Finally, in **Espejo-Garcia et al., (2020)**, authors used transfer learning using agricultural datasets for pre-training neural networks, and afterwards, they fine-tuned the networks for classifying 4 species extracted from the **Eden Platform**. Their maximum performance was 99.54% by using the **Xception** architecture.\n", "\n", "It is important to note that in this notebook a technique integrating neural-network feature-extraction and \"traditional\" machine learning algorithms is used. This technique was used in **Kounalakis et al., (2019)** and **Espejo-Garcia et al., (2020)**, and represents an extension over the previous Eden notebooks:\n", "1. https://github.com/Eden-Library-AI/eden_library_notebooks/blob/master/image_classification/weeds_identification-transfer_learning-1.ipynb\n", "2. https://github.com/Eden-Library-AI/eden_library_notebooks/blob/master/image_classification/weeds_identification-transfer_learning-2.ipynb\n", "3. https://github.com/Eden-Library-AI/eden_library_notebooks/blob/master/image_classification/weeds_identification-transfer_learning-3.ipynb\n", "4. https://github.com/Eden-Library-AI/eden_library_notebooks/blob/master/image_classification/weeds_identification-transfer_learning-4.ipynb" ] }, { "cell_type": "markdown", "metadata": { "id": "vyIF86zz0hfu" }, "source": [ "### Library Imports" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "executionInfo": { "elapsed": 2928, "status": "ok", "timestamp": 1622411696291, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "hj9NVykKZM_z" }, "outputs": [], "source": [ "import warnings\n", "\n", "warnings.filterwarnings(\"ignore\")\n", "\n", "import numpy as np\n", "import cv2\n", "import os\n", "import csv\n", "import gc\n", "import random\n", "import matplotlib.pyplot as plt\n", "\n", "from tqdm import tqdm\n", "from glob import glob\n", "from pathlib import Path\n", "\n", "import tensorflow as tf\n", "from tensorflow.keras.applications import *\n", "from tensorflow.keras import layers\n", "from tensorflow.keras.models import Model\n", "from tensorflow.keras.optimizers import Adam\n", "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint\n", "from tensorflow.keras.callbacks import ReduceLROnPlateau\n", "import tensorflow.keras.backend as K\n", "\n", "from sklearn.model_selection import train_test_split" ] }, { "cell_type": "markdown", "metadata": { "id": "uuV03GcT0hfv" }, "source": [ "### Auxiliary functions\n", "Check the docstrings for more information." ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "executionInfo": { "elapsed": 198, "status": "ok", "timestamp": 1622411697214, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "a0AftNp7euWr" }, "outputs": [], "source": [ "def plot_sample(X):\n", " \"\"\"\n", " Given the array of images , it plots a random subsample of 9 of them.\n", "\n", " Parameters:\n", " X (ndarray): The array with all the images.\n", " \"\"\"\n", " nb_rows = 3\n", " nb_cols = 3\n", " fig, axs = plt.subplots(nb_rows, nb_cols, figsize=(6, 6))\n", "\n", " for i in range(0, nb_rows):\n", " for j in range(0, nb_cols):\n", " axs[i, j].xaxis.set_ticklabels([])\n", " axs[i, j].yaxis.set_ticklabels([])\n", " axs[i, j].imshow(X[random.randint(0, X.shape[0] - 1)])\n", "\n", "\n", "def read_data(path_list, im_size=(224, 224)):\n", " \"\"\"\n", " Given the list of paths where the images are stored ,\n", " and the size for image decimation , it returns 2 Numpy Arrays\n", " with the images and labels.\n", "\n", " Parameters:\n", " path_list (List[String]): The list of paths to the images.\n", " im_size (Tuple): The height and width values.\n", "\n", " Returns:\n", " X (ndarray): Images\n", " y (ndarray): Labels\n", " \"\"\"\n", " X = []\n", " y = []\n", "\n", " # Extract filenames of the datasets we ingest and create a label dictionary\n", " tag2idx = {tag.split(os.path.sep)[-1]: i for i, tag in enumerate(path_list)}\n", "\n", " for path in path_list:\n", " for im_file in tqdm(glob(path + \"*/*\")): # Read all files in path\n", " try:\n", " # os.path.separator is OS agnostic (either '/' or '\\'),[-2] to grab folder name\n", " label = im_file.split(os.path.sep)[-2]\n", " im = cv2.imread(im_file, cv2.IMREAD_COLOR)\n", " # By default OpenCV reads with BGR format, convert to RGB\n", " im = cv2.cvtColor(im, cv2.COLOR_BGR2RGB)\n", " # Resize to appropriate dimensions\n", " im = cv2.resize(im, im_size, interpolation=cv2.INTER_AREA)\n", " X.append(im)\n", " y.append(tag2idx[label]) # Append the label name to y\n", " except Exception as e:\n", " # In case annotations or metadata are found\n", " print(\"Not a picture\")\n", "\n", " X = np.array(X) # Convert list to numpy array.\n", " y = np.eye(len(np.unique(y)))[y].astype(np.uint8)\n", "\n", " return X, y\n", "\n", "\n", "def get_callbacks(weights_file, patience, lr_factor):\n", " \"\"\"\n", " Callbacks are used for saving the best weights and early stopping.\n", " Given some configuration parameters, it creates the callbacks that\n", " will be used by Keras after each epoch.\n", "\n", " Parameters:\n", " weights_file (String): File name for saving the best model weights.\n", " patience (Integer): Number of epochs without improvement to wait.\n", " lr_factor: Factor for reducing the learning rate when performance\n", " is not improving.\n", "\n", " Returns:\n", " callbacks (List[Callbacks]): Configured callback ready for using.\n", " \"\"\"\n", " return [\n", " # Only save weights that correspond to the maximum validation accuracy\n", " ModelCheckpoint(\n", " filepath=weights_file,\n", " monitor=\"val_accuracy\",\n", " mode=\"max\",\n", " save_best_only=True,\n", " save_weights_only=True,\n", " ),\n", " # If val_loss does not improve after a number of epochs (patience),\n", " # training will stop to avoid overfitting.\n", " EarlyStopping(monitor=\"val_loss\", mode=\"min\", patience=patience, verbose=1),\n", " # Learning rate is reduced by 'lr_factor' if val_loss stagnates\n", " # for a number of epochs (patience).\n", " # ReduceLROnPlateau(monitor=\"val_loss\", mode=\"min\",\n", " # factor=lr_factor, min_lr=1e-6, patience=patience//2, verbose=1 )\n", " ]\n", "\n", "\n", "def plot_performances(performances):\n", " \"\"\"\n", " Given the list of performances (validation accuracies) and method-name ,\n", " it plots how the validation accuracy progressed during the training/validation process.\n", "\n", " Parameters:\n", " performances (List[Tuple]): The list of method-performance tuples.\n", " \"\"\"\n", " plt.title(\"Validation Accuracy vs. Number of Training Epochs\")\n", " plt.xlabel(\"Training Epochs\")\n", " plt.ylabel(\"Validation Accuracy\")\n", " for performance in performances:\n", " plt.plot(\n", " range(1, len(performance[1]) + 1), performance[1], label=performance[0]\n", " )\n", " plt.ylim((0.5, 1.05))\n", " plt.xticks(np.arange(1, NUM_EPOCHS + 1, 1.0))\n", " plt.legend()\n", " plt.show()" ] }, { "cell_type": "markdown", "metadata": { "id": "gFfTG8Ga_Ibb" }, "source": [ "### Experimental Constants" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "executionInfo": { "elapsed": 196, "status": "ok", "timestamp": 1622412905314, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "5r7CAF-c9ASV" }, "outputs": [], "source": [ "INPUT_SHAPE = (224, 224, 3)\n", "IM_SIZE = (224, 224)\n", "NUM_EPOCHS = 20\n", "BATCH_SIZE = 4\n", "TEST_SPLIT = 0.2\n", "VAL_SPLIT = 0.2\n", "RANDOM_STATE = 2021\n", "WEIGHTS_FILE = \"weights.h5\" # File to store updated weights\n", "\n", "# Eden datasets we will work on\n", "PATH_LIST = [\n", " \"Cotton-100619-Healthy-zz-V1-20210225102300\",\n", " \"Black nightsade-220519-Weed-zz-V1-20210225102034\",\n", " \"Tomato-240519-Healthy-zz-V1-20210225103740\",\n", " \"Velvet leaf-220519-Weed-zz-V1-20210225104123\",\n", "]" ] }, { "cell_type": "markdown", "metadata": { "id": "IPE0KaIa_PHw" }, "source": [ "### Loading images and Data Loaders" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "executionInfo": { "elapsed": 114969, "status": "ok", "timestamp": 1622411824220, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "45J4jLyYZzQV", "outputId": "4a387c42-3acf-48cf-ba45-5c9d8aa9b28e" }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|██████████| 47/47 [00:11<00:00, 4.15it/s]\n", "100%|██████████| 123/123 [00:28<00:00, 4.26it/s]\n", "100%|██████████| 201/201 [01:09<00:00, 2.91it/s]\n", "100%|██████████| 129/129 [00:31<00:00, 4.11it/s]\n" ] } ], "source": [ "i = 0\n", "for path in PATH_LIST:\n", " # Define paths in an OS agnostic way.\n", " PATH_LIST[i] = str(\n", " Path(Path.cwd()).parents[0].joinpath(\"eden_library_datasets\").joinpath(path)\n", " )\n", " i += 1\n", "X, y = read_data(PATH_LIST, IM_SIZE)" ] }, { "cell_type": "markdown", "metadata": { "id": "iZ6lZksO_U2g" }, "source": [ "### Plot sample images" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 361 }, "executionInfo": { "elapsed": 1611, "status": "ok", "timestamp": 1622411863999, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "grzjRE3hClkH", "outputId": "1c777845-14dc-4d36-972b-72061b52c50f" }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plot_sample(X)" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "executionInfo": { "elapsed": 192, "status": "ok", "timestamp": 1622412363557, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "qeZd9vwYjgMK" }, "outputs": [], "source": [ "def get_architecture(y, mobilenet_version, as_feature_extractor):\n", " \"\"\"\n", " This function returns a pre-trained Tensorflow MobileNet architecture.\n", " Versions 2 and 3-Large are available.\n", "\n", " Parameters:\n", " y (ndarray): Array with labels. It is used for computing the number of classes.\n", " mobilenet_version (string): which MobileNet version to import (2 or 3 Large).\n", " as_feature_extractor (Boolean): Whether to fine-tune the \"convolutional\" layers.\n", "\n", " Returns:\n", " model (Model): MobileNet architecture ready for training.\n", " \"\"\"\n", "\n", " if mobilenet_version == \"2\":\n", " feature_extractor = MobileNetV2(\n", " weights=\"imagenet\", # Load weights pre-trained on ImageNet\n", " include_top=False, # Do not include the ImageNet classifier at the top\n", " input_shape=INPUT_SHAPE,\n", " )\n", " elif mobilenet_version == \"3\":\n", " feature_extractor = MobileNetV3Large(\n", " weights=\"imagenet\", # Load weights pre-trained on ImageNet\n", " include_top=False, # Do not include the ImageNet classifier at the top\n", " input_shape=INPUT_SHAPE,\n", " )\n", "\n", " # Freeze the base_model,we don't want to update initial weights\n", " if as_feature_extractor:\n", " feature_extractor.trainable = False\n", " else:\n", " feature_extractor.trainable = True # Not necessary\n", "\n", " # Create new model on top.\n", " x = layers.GlobalAveragePooling2D(name=\"pool\")(\n", " feature_extractor.output\n", " ) # Flattening layer\n", " x = layers.Dense(units=64, activation=\"relu\")(x) # Fully connected layer\n", " # Create a Classifier with shape=number_of_training_classes.\n", " x = layers.Dropout(0.3)(x) # Regularize with dropout.\n", " out = layers.Dense(units=y.shape[1], activation=\"softmax\")(x)\n", " model = Model(feature_extractor.input, out) # Final model\n", "\n", " # Defining base learning rate for Adam optimizer\n", " base_learning_rate = 1e-4\n", " model.compile(\n", " loss=\"categorical_crossentropy\",\n", " optimizer=tf.keras.optimizers.Adam(lr=base_learning_rate),\n", " metrics=[\"accuracy\"],\n", " )\n", "\n", " return model" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Training vs Fine-Tuning Models\n", "When using transfer learning, there are several strategies for reusing the weights or the knowledge learned on another task. Here 2 strategies are shown and compared:\n", "\n", "1. **Feature Extraction**: The weights of the pre-trained architecture are not updated (the layers are frozen). This way, the architecture is used as a feature extractor while the last layer or classifier is trained to make sense of these features in the new problem.\n", "2. **Fine-Tuning**: The weights of the pre-trained architecture are updated. The number of layers that are trained depends on the specific problem addressed and many times there is not a predefined best option." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "executionInfo": { "elapsed": 550, "status": "ok", "timestamp": 1622412383875, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "17nOWwEfZJzU" }, "outputs": [], "source": [ "X_prep = tf.keras.applications.mobilenet_v3.preprocess_input(X)\n", "\n", "X_train, X_test, y_train, y_test = train_test_split(\n", " X_prep,\n", " y, \n", " test_size=TEST_SPLIT, \n", " shuffle=True, \n", " stratify=y, \n", " random_state=RANDOM_STATE\n", ")\n", "\n", "X_train, X_val, y_train, y_val = train_test_split(\n", " X_train,\n", " y_train,\n", " test_size=VAL_SPLIT,\n", " shuffle=True,\n", " stratify=y_train,\n", " random_state=RANDOM_STATE,\n", ")" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "executionInfo": { "elapsed": 92585, "status": "ok", "timestamp": 1622412476457, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "9pZ8X7jwZK96", "outputId": "157aba05-0e74-457e-8ee4-f7f749a400db" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/20\n", "80/80 [==============================] - 6s 56ms/step - loss: 1.3755 - accuracy: 0.4344 - val_loss: 1.0574 - val_accuracy: 0.6500\n", "Epoch 2/20\n", "80/80 [==============================] - 4s 49ms/step - loss: 1.0204 - accuracy: 0.6031 - val_loss: 0.9086 - val_accuracy: 0.6250\n", "Epoch 3/20\n", "80/80 [==============================] - 4s 51ms/step - loss: 0.9348 - accuracy: 0.6406 - val_loss: 0.8375 - val_accuracy: 0.6750\n", "Epoch 4/20\n", "80/80 [==============================] - 4s 48ms/step - loss: 0.7829 - accuracy: 0.7125 - val_loss: 0.7806 - val_accuracy: 0.6875\n", "Epoch 5/20\n", "80/80 [==============================] - 4s 50ms/step - loss: 0.7482 - accuracy: 0.7281 - val_loss: 0.7264 - val_accuracy: 0.6875\n", "Epoch 6/20\n", "80/80 [==============================] - 4s 49ms/step - loss: 0.6840 - accuracy: 0.7656 - val_loss: 0.7084 - val_accuracy: 0.7125\n", "Epoch 7/20\n", "80/80 [==============================] - 4s 49ms/step - loss: 0.6588 - accuracy: 0.7750 - val_loss: 0.6626 - val_accuracy: 0.6875\n", "Epoch 8/20\n", "80/80 [==============================] - 4s 52ms/step - loss: 0.6090 - accuracy: 0.7750 - val_loss: 0.6398 - val_accuracy: 0.7500\n", "Epoch 9/20\n", "80/80 [==============================] - 4s 51ms/step - loss: 0.5610 - accuracy: 0.8188 - val_loss: 0.6079 - val_accuracy: 0.7250\n", "Epoch 10/20\n", "80/80 [==============================] - 4s 50ms/step - loss: 0.5465 - accuracy: 0.8281 - val_loss: 0.5910 - val_accuracy: 0.7625\n", "Epoch 11/20\n", "80/80 [==============================] - 4s 47ms/step - loss: 0.5160 - accuracy: 0.8406 - val_loss: 0.5652 - val_accuracy: 0.7125\n", "Epoch 12/20\n", "80/80 [==============================] - 4s 47ms/step - loss: 0.4711 - accuracy: 0.8281 - val_loss: 0.5502 - val_accuracy: 0.7750\n", "Epoch 13/20\n", "80/80 [==============================] - 4s 47ms/step - loss: 0.4733 - accuracy: 0.8250 - val_loss: 0.5249 - val_accuracy: 0.8000\n", "Epoch 14/20\n", "80/80 [==============================] - 4s 51ms/step - loss: 0.4357 - accuracy: 0.8687 - val_loss: 0.5268 - val_accuracy: 0.7500\n", "Epoch 15/20\n", "80/80 [==============================] - 4s 54ms/step - loss: 0.4189 - accuracy: 0.8531 - val_loss: 0.4959 - val_accuracy: 0.8375\n", "Epoch 16/20\n", "80/80 [==============================] - 4s 50ms/step - loss: 0.3850 - accuracy: 0.8844 - val_loss: 0.4907 - val_accuracy: 0.8125\n", "Epoch 17/20\n", "80/80 [==============================] - 5s 60ms/step - loss: 0.3781 - accuracy: 0.8844 - val_loss: 0.4889 - val_accuracy: 0.8375\n", "Epoch 18/20\n", "80/80 [==============================] - 4s 48ms/step - loss: 0.3889 - accuracy: 0.8656 - val_loss: 0.4670 - val_accuracy: 0.8625\n", "Epoch 19/20\n", "80/80 [==============================] - 4s 51ms/step - loss: 0.3600 - accuracy: 0.8938 - val_loss: 0.4577 - val_accuracy: 0.8375\n", "Epoch 20/20\n", "80/80 [==============================] - 4s 55ms/step - loss: 0.3235 - accuracy: 0.9031 - val_loss: 0.4420 - val_accuracy: 0.8500\n", "************************************************************************\n", "Final MobileNetV2 (Feature Extraction) Accuracy: 0.8700000047683716\n", "************************************************************************\n" ] } ], "source": [ "model = get_architecture(y, mobilenet_version=\"2\", as_feature_extractor=True)\n", "\n", "history_v2_fe = model.fit(\n", " X_train,\n", " y_train,\n", " batch_size=BATCH_SIZE,\n", " epochs=NUM_EPOCHS,\n", " validation_data=(X_val, y_val),\n", " callbacks=get_callbacks(WEIGHTS_FILE, NUM_EPOCHS // 2, 0.25),\n", ")\n", "\n", "model.load_weights(WEIGHTS_FILE)\n", "final_accuracy = model.evaluate(X_test, y_test, batch_size=1, verbose=0)[1]\n", "print(\"*\" * 72)\n", "print(f\"Final MobileNetV2 (Feature Extraction) Accuracy: {final_accuracy}\")\n", "print(\"*\" * 72)" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/20\n", "80/80 [==============================] - 25s 287ms/step - loss: 0.9497 - accuracy: 0.5938 - val_loss: 1.5658 - val_accuracy: 0.3750\n", "Epoch 2/20\n", "80/80 [==============================] - 26s 321ms/step - loss: 0.4912 - accuracy: 0.8281 - val_loss: 1.3171 - val_accuracy: 0.3625\n", "Epoch 3/20\n", "80/80 [==============================] - 23s 283ms/step - loss: 0.2738 - accuracy: 0.9031 - val_loss: 1.3349 - val_accuracy: 0.4625\n", "Epoch 4/20\n", "80/80 [==============================] - 23s 294ms/step - loss: 0.2752 - accuracy: 0.9156 - val_loss: 1.4152 - val_accuracy: 0.4750\n", "Epoch 5/20\n", "80/80 [==============================] - 22s 281ms/step - loss: 0.2558 - accuracy: 0.9094 - val_loss: 1.7777 - val_accuracy: 0.5250\n", "Epoch 6/20\n", "80/80 [==============================] - 21s 268ms/step - loss: 0.1960 - accuracy: 0.9312 - val_loss: 1.8518 - val_accuracy: 0.5625\n", "Epoch 7/20\n", "80/80 [==============================] - 21s 267ms/step - loss: 0.1309 - accuracy: 0.9625 - val_loss: 1.5597 - val_accuracy: 0.6000\n", "Epoch 8/20\n", "80/80 [==============================] - 22s 281ms/step - loss: 0.1095 - accuracy: 0.9563 - val_loss: 2.1635 - val_accuracy: 0.6000\n", "Epoch 9/20\n", "80/80 [==============================] - 23s 287ms/step - loss: 0.0710 - accuracy: 0.9750 - val_loss: 1.4020 - val_accuracy: 0.7000\n", "Epoch 10/20\n", "80/80 [==============================] - 23s 286ms/step - loss: 0.0983 - accuracy: 0.9625 - val_loss: 1.5237 - val_accuracy: 0.6625\n", "Epoch 11/20\n", "80/80 [==============================] - 24s 298ms/step - loss: 0.0977 - accuracy: 0.9688 - val_loss: 0.9035 - val_accuracy: 0.7750\n", "Epoch 12/20\n", "80/80 [==============================] - 29s 359ms/step - loss: 0.0797 - accuracy: 0.9750 - val_loss: 0.1304 - val_accuracy: 0.9250\n", "Epoch 13/20\n", "80/80 [==============================] - 38s 477ms/step - loss: 0.1426 - accuracy: 0.9594 - val_loss: 0.2755 - val_accuracy: 0.9125\n", "Epoch 14/20\n", "80/80 [==============================] - 41s 517ms/step - loss: 0.1187 - accuracy: 0.9625 - val_loss: 0.1243 - val_accuracy: 0.9500\n", "Epoch 15/20\n", "80/80 [==============================] - 44s 552ms/step - loss: 0.1172 - accuracy: 0.9656 - val_loss: 0.3091 - val_accuracy: 0.9000\n", "Epoch 16/20\n", "80/80 [==============================] - 44s 555ms/step - loss: 0.0747 - accuracy: 0.9750 - val_loss: 0.2560 - val_accuracy: 0.9000\n", "Epoch 17/20\n", "80/80 [==============================] - 36s 450ms/step - loss: 0.1234 - accuracy: 0.9531 - val_loss: 0.0801 - val_accuracy: 0.9750\n", "Epoch 18/20\n", "80/80 [==============================] - 24s 299ms/step - loss: 0.0654 - accuracy: 0.9906 - val_loss: 0.0747 - val_accuracy: 0.9750\n", "Epoch 19/20\n", "80/80 [==============================] - 23s 284ms/step - loss: 0.0615 - accuracy: 0.9781 - val_loss: 0.2768 - val_accuracy: 0.9500\n", "Epoch 20/20\n", "80/80 [==============================] - 23s 282ms/step - loss: 0.0204 - accuracy: 0.9937 - val_loss: 0.2577 - val_accuracy: 0.9625\n", "************************************************************************\n", "Final MobileNetV2 (Fine-Tuning) Accuracy: 0.9700000286102295\n", "************************************************************************\n" ] } ], "source": [ "model = get_architecture(y, mobilenet_version=\"2\", as_feature_extractor=False)\n", "\n", "history_v2_ft = model.fit(\n", " X_train,\n", " y_train,\n", " batch_size=BATCH_SIZE,\n", " epochs=NUM_EPOCHS,\n", " validation_data=(X_val, y_val),\n", " callbacks=get_callbacks(WEIGHTS_FILE, NUM_EPOCHS // 2, 0.25),\n", ")\n", "model.load_weights(WEIGHTS_FILE)\n", "final_accuracy = model.evaluate(X_test, y_test, batch_size=1, verbose=0)[1]\n", "print(\"*\" * 72)\n", "print(f\"Final MobileNetV2 (Fine-Tuning) Accuracy: {final_accuracy}\")\n", "print(\"*\" * 72)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "executionInfo": { "elapsed": 186, "status": "ok", "timestamp": 1622412625352, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "qLZT3h9oBeTy" }, "outputs": [], "source": [ "X_prep = mobilenet_v3.preprocess_input(X)\n", "\n", "X_train, X_test, y_train, y_test = train_test_split(\n", " X_prep, \n", " y, \n", " test_size=TEST_SPLIT, \n", " shuffle=True, \n", " stratify=y, \n", " random_state=RANDOM_STATE\n", ")\n", "\n", "X_train, X_val, y_train, y_val = train_test_split(\n", " X_train,\n", " y_train,\n", " test_size=VAL_SPLIT,\n", " shuffle=True,\n", " stratify=y_train,\n", " random_state=RANDOM_STATE,\n", ")" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "executionInfo": { "elapsed": 98390, "status": "ok", "timestamp": 1622412724619, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "Q1Qrk5DuCBln", "outputId": "698fa539-ac36-4baa-8d09-60e6ccdf7d21" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/20\n", "80/80 [==============================] - 7s 60ms/step - loss: 1.1392 - accuracy: 0.5344 - val_loss: 0.8491 - val_accuracy: 0.6500\n", "Epoch 2/20\n", "80/80 [==============================] - 4s 51ms/step - loss: 0.6999 - accuracy: 0.7719 - val_loss: 0.6801 - val_accuracy: 0.7250\n", "Epoch 3/20\n", "80/80 [==============================] - 4s 51ms/step - loss: 0.5378 - accuracy: 0.8313 - val_loss: 0.5307 - val_accuracy: 0.8375\n", "Epoch 4/20\n", "80/80 [==============================] - 4s 51ms/step - loss: 0.3953 - accuracy: 0.8844 - val_loss: 0.4520 - val_accuracy: 0.8625\n", "Epoch 5/20\n", "80/80 [==============================] - 4s 51ms/step - loss: 0.3484 - accuracy: 0.9187 - val_loss: 0.3977 - val_accuracy: 0.8875\n", "Epoch 6/20\n", "80/80 [==============================] - 4s 54ms/step - loss: 0.3217 - accuracy: 0.9062 - val_loss: 0.3559 - val_accuracy: 0.9000\n", "Epoch 7/20\n", "80/80 [==============================] - 4s 52ms/step - loss: 0.2609 - accuracy: 0.9281 - val_loss: 0.3151 - val_accuracy: 0.9125\n", "Epoch 8/20\n", "80/80 [==============================] - 5s 62ms/step - loss: 0.2472 - accuracy: 0.9438 - val_loss: 0.3016 - val_accuracy: 0.9125\n", "Epoch 9/20\n", "80/80 [==============================] - 5s 58ms/step - loss: 0.1786 - accuracy: 0.9656 - val_loss: 0.2525 - val_accuracy: 0.9500\n", "Epoch 10/20\n", "80/80 [==============================] - 5s 57ms/step - loss: 0.1870 - accuracy: 0.9594 - val_loss: 0.2317 - val_accuracy: 0.9625\n", "Epoch 11/20\n", "80/80 [==============================] - 4s 55ms/step - loss: 0.1509 - accuracy: 0.9688 - val_loss: 0.2086 - val_accuracy: 0.9625\n", "Epoch 12/20\n", "80/80 [==============================] - 5s 61ms/step - loss: 0.1337 - accuracy: 0.9719 - val_loss: 0.1979 - val_accuracy: 0.9625\n", "Epoch 13/20\n", "80/80 [==============================] - 4s 54ms/step - loss: 0.1424 - accuracy: 0.9688 - val_loss: 0.1959 - val_accuracy: 0.9625\n", "Epoch 14/20\n", "80/80 [==============================] - 4s 55ms/step - loss: 0.1261 - accuracy: 0.9688 - val_loss: 0.1924 - val_accuracy: 0.9500\n", "Epoch 15/20\n", "80/80 [==============================] - 4s 54ms/step - loss: 0.0992 - accuracy: 0.9812 - val_loss: 0.1704 - val_accuracy: 0.9750\n", "Epoch 16/20\n", "80/80 [==============================] - 4s 54ms/step - loss: 0.1104 - accuracy: 0.9844 - val_loss: 0.1550 - val_accuracy: 0.9750\n", "Epoch 17/20\n", "80/80 [==============================] - 4s 55ms/step - loss: 0.0818 - accuracy: 0.9937 - val_loss: 0.1466 - val_accuracy: 0.9750\n", "Epoch 18/20\n", "80/80 [==============================] - 4s 55ms/step - loss: 0.0799 - accuracy: 0.9844 - val_loss: 0.1485 - val_accuracy: 0.9750\n", "Epoch 19/20\n", "80/80 [==============================] - 4s 54ms/step - loss: 0.0724 - accuracy: 0.9844 - val_loss: 0.1360 - val_accuracy: 0.9750\n", "Epoch 20/20\n", "80/80 [==============================] - 4s 55ms/step - loss: 0.0704 - accuracy: 0.9875 - val_loss: 0.1309 - val_accuracy: 0.9750\n", "************************************************************************\n", "Final MobileNetV3 (Feature Extraction) Accuracy: 0.9599999785423279\n", "************************************************************************\n" ] } ], "source": [ "model = get_architecture(y, mobilenet_version=\"3\", as_feature_extractor=True)\n", "\n", "history_v3_fe = model.fit(\n", " X_train,\n", " y_train,\n", " batch_size=BATCH_SIZE,\n", " epochs=NUM_EPOCHS,\n", " validation_data=(X_val, y_val),\n", " callbacks=get_callbacks(WEIGHTS_FILE, NUM_EPOCHS // 2, 0.25),\n", ")\n", "\n", "model.load_weights(WEIGHTS_FILE)\n", "final_accuracy = model.evaluate(X_test, y_test, batch_size=1, verbose=0)[1]\n", "print(\"*\" * 72)\n", "print(f\"Final MobileNetV3 (Feature Extraction) Accuracy: {final_accuracy}\")\n", "print(\"*\" * 72)" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/20\n", "80/80 [==============================] - 23s 254ms/step - loss: 1.1188 - accuracy: 0.5469 - val_loss: 0.9398 - val_accuracy: 0.6000\n", "Epoch 2/20\n", "80/80 [==============================] - 20s 250ms/step - loss: 0.6907 - accuracy: 0.7188 - val_loss: 0.6367 - val_accuracy: 0.7625\n", "Epoch 3/20\n", "80/80 [==============================] - 21s 258ms/step - loss: 0.3004 - accuracy: 0.8969 - val_loss: 0.4886 - val_accuracy: 0.8000\n", "Epoch 4/20\n", "80/80 [==============================] - 21s 260ms/step - loss: 0.2457 - accuracy: 0.9187 - val_loss: 0.3991 - val_accuracy: 0.8375\n", "Epoch 5/20\n", "80/80 [==============================] - 21s 266ms/step - loss: 0.1495 - accuracy: 0.9563 - val_loss: 0.1898 - val_accuracy: 0.9500\n", "Epoch 6/20\n", "80/80 [==============================] - 21s 262ms/step - loss: 0.0669 - accuracy: 0.9781 - val_loss: 0.1758 - val_accuracy: 0.9375\n", "Epoch 7/20\n", "80/80 [==============================] - 26s 323ms/step - loss: 0.1111 - accuracy: 0.9688 - val_loss: 0.1496 - val_accuracy: 0.9625\n", "Epoch 8/20\n", "80/80 [==============================] - 27s 337ms/step - loss: 0.0752 - accuracy: 0.9781 - val_loss: 0.1734 - val_accuracy: 0.9625\n", "Epoch 9/20\n", "80/80 [==============================] - 28s 346ms/step - loss: 0.1059 - accuracy: 0.9594 - val_loss: 0.0577 - val_accuracy: 0.9875\n", "Epoch 10/20\n", "80/80 [==============================] - 29s 357ms/step - loss: 0.0827 - accuracy: 0.9781 - val_loss: 0.0624 - val_accuracy: 0.9875\n", "Epoch 11/20\n", "80/80 [==============================] - 25s 318ms/step - loss: 0.0443 - accuracy: 0.9906 - val_loss: 0.0863 - val_accuracy: 0.9875\n", "Epoch 12/20\n", "80/80 [==============================] - 23s 287ms/step - loss: 0.0349 - accuracy: 0.9906 - val_loss: 0.1065 - val_accuracy: 0.9875\n", "Epoch 13/20\n", "80/80 [==============================] - 22s 278ms/step - loss: 0.0177 - accuracy: 0.9969 - val_loss: 0.1084 - val_accuracy: 0.9875\n", "Epoch 14/20\n", "80/80 [==============================] - 23s 286ms/step - loss: 0.0513 - accuracy: 0.9844 - val_loss: 0.1308 - val_accuracy: 0.9875\n", "Epoch 15/20\n", "80/80 [==============================] - 22s 272ms/step - loss: 0.0257 - accuracy: 0.9906 - val_loss: 0.1318 - val_accuracy: 0.9875\n", "Epoch 16/20\n", "80/80 [==============================] - 22s 274ms/step - loss: 0.0292 - accuracy: 0.9906 - val_loss: 0.1025 - val_accuracy: 0.9875\n", "Epoch 17/20\n", "80/80 [==============================] - 22s 274ms/step - loss: 0.0201 - accuracy: 0.9969 - val_loss: 0.0482 - val_accuracy: 0.9875\n", "Epoch 18/20\n", "80/80 [==============================] - 22s 281ms/step - loss: 0.0373 - accuracy: 0.9844 - val_loss: 0.0482 - val_accuracy: 0.9875\n", "Epoch 19/20\n", "80/80 [==============================] - 23s 286ms/step - loss: 0.0268 - accuracy: 0.9906 - val_loss: 0.0394 - val_accuracy: 0.9750\n", "Epoch 20/20\n", "80/80 [==============================] - 23s 282ms/step - loss: 0.0510 - accuracy: 0.9812 - val_loss: 0.0542 - val_accuracy: 0.9625\n", "************************************************************************\n", "Final MobileNetV3 (Fine Tuning) Accuracy: 1.0\n", "************************************************************************\n" ] } ], "source": [ "model = get_architecture(y, mobilenet_version=\"3\", as_feature_extractor=False)\n", "\n", "history_v3_ft = model.fit(\n", " X_train,\n", " y_train,\n", " batch_size=BATCH_SIZE,\n", " epochs=NUM_EPOCHS,\n", " validation_data=(X_val, y_val),\n", " callbacks=get_callbacks(WEIGHTS_FILE, NUM_EPOCHS // 2, 0.25),\n", ")\n", "\n", "model.load_weights(WEIGHTS_FILE)\n", "final_accuracy = model.evaluate(X_test, y_test, batch_size=1, verbose=0)[1]\n", "print(\"*\" * 72)\n", "print(f\"Final MobileNetV3 (Fine Tuning) Accuracy: {final_accuracy}\")\n", "print(\"*\" * 72)" ] }, { "cell_type": "markdown", "metadata": { "id": "I0CwQLvF0hfz" }, "source": [ "### Plotting model performances" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "executionInfo": { "elapsed": 3, "status": "ok", "timestamp": 1622413080523, "user": { "displayName": "borja Espejo", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gigx3Qu0obXmM3V-jNhAmhVoUdNxK5dupK93bPfKw=s64", "userId": "14834259721458229847" }, "user_tz": -120 }, "id": "PLJnrlqSr-DG" }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plot_performances([\n", " (\"MobileNetV2-Fine Tuning\", history_v2_ft.history[\"val_accuracy\"]),\n", " (\"MobileNetV2-Feature Extractor\", history_v2_fe.history[\"val_accuracy\"]),\n", " (\"MobileNetV3-Fine Tuning\", history_v3_ft.history[\"val_accuracy\"]),\n", " (\"MobileNetV3-Feature Extractor\", history_v3_fe.history[\"val_accuracy\"]),\n", "])" ] }, { "cell_type": "markdown", "metadata": { "id": "dqhVkpKHbfCy" }, "source": [ "## Possible Extensions\n", "1. Use a different pre-trained network (e.g.: MobileNetV3Small).\n", "2. Try a different training approach where pre-trained weights are not loaded.\n", "3. Try different learning rates (0.0001 in this notebook) or even other optimization algorithms (e.g.: SGD).\n", "4. Try different epochs and batch sizes." ] }, { "cell_type": "markdown", "metadata": { "id": "b3jq6zle0hf0" }, "source": [ "## Bibliography\n", "Bengio, Y., 2012. Deep Learning of Representations for Unsupervised and Transfer Learning. In: Journal of Machine Learning Research; 17–37.\n", "\n", "Wang, G., Sun, Y., Wang, J., (2017). Automatic Image-Based Plant Disease Severity Estimation Using Deep Learning. Computational Intelligence and Neuroscience; 2017:8.\n", "\n", "Mehdipour-Ghazi, M., Yanikoglu, B.A., & Aptoula, E. (2017). Plant identification using deep neural networks via optimization of transfer learning parameters. Neurocomputing, 235, 228-235.\n", "\n", "Suh, H.K., IJsselmuiden, J., Hofstee, J.W., van Henten, E.J., (2018). Transfer learning for the classification of sugar beet and volunteer potato under field conditions. Biosystems Engineering; 174:50–65.\n", "\n", "Kounalakis T., Triantafyllidis G. A., Nalpantidis L., (2019). Deep learning-based visual recognition of rumex for robotic precision farming. Computers and Electronics in Agriculture.\n", "\n", "Too, E.C., Yujian, L., Njuki, S., & Ying-chun, L. (2019). A comparative study of fine-tuning deep learning models for plant disease identification. Comput. Electron. Agric., 161, 272-279.\n", "\n", "Espejo-Garcia, B., Mylonas, N., Athanasakos, L., & Fountas, S., (2020). Improving\n", "Weeds Identification with a Repository of Agricultural Pre-trained Deep Neural\n", "Networks. Computers and Electronics in Agriculture; 175 (August).\n", "\n", "Sandler, M., Howard, A.G., Zhu, M., Zhmoginov, A., & Chen, L. (2018). MobileNetV2: Inverted Residuals and Linear Bottlenecks. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4510-4520.\n", "\n", "Howard, A.G., Sandler, M., Chu, G., Chen, L., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., Le, Q.V., & Adam, H. (2019). Searching for MobileNetV3. 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 1314-1324.\n", "\n", "https://www.tensorflow.org/api_docs/python/tf/keras/applications" ] } ], "metadata": { "colab": { "collapsed_sections": [], "name": "weeds_identification-transfer_learning-5.ipynb", "provenance": [] }, "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.16" } }, "nbformat": 4, "nbformat_minor": 4 }