{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"_kg_hide-input":true},"cell_type":"code","source":"import numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport seaborn as sns\nimport matplotlib.pyplot as plt\nimport cv2\nfrom sklearn import preprocessing\nfrom tqdm import tqdm\nimport os","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-input":true},"cell_type":"code","source":"train_df = pd.read_csv('../input/plant-pathology-2021-fgvc8/train.csv')\ntrain_df","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-input":true},"cell_type":"code","source":"train_df.labels.value_counts().to_frame().style.background_gradient(cmap=\"plasma\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-input":true},"cell_type":"code","source":"plt.figure(figsize=(8, 4))\nsns.countplot(y=\"labels\", data=train_df)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"As we can see, the dataset has a fairly large imbalance."},{"metadata":{},"cell_type":"markdown","source":"Let's label encode the classes"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"# label_encoder object knows how to understand word labels.\nlabel_encoder = preprocessing.LabelEncoder()\n \n# Encode labels in column 'species'.\ntrain_df[\"labels_code\"]= label_encoder.fit_transform(train_df[[\"labels\"]])\ntrain_df","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Let's take a look at the dimensions of the first 300 images \n\nAs you can see below, all images are of different sizes."},{"metadata":{"trusted":true,"_kg_hide-input":true},"cell_type":"code","source":"BASE_DIR = \"../input/plant-pathology-2021-fgvc8\"\n\nimg_shapes = {}\nfor image_name in tqdm(os.listdir(os.path.join(BASE_DIR, \"train_images\"))[:300]):\n image = cv2.imread(os.path.join(BASE_DIR, \"train_images\", image_name))\n img_shapes[image.shape] = img_shapes.get(image.shape, 0) + 1\n\nprint(img_shapes)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-input":true},"cell_type":"code","source":"def visualize_batch(image_ids, labels):\n plt.figure(figsize=(16, 12))\n \n for ind, (image_id, label) in enumerate(zip(image_ids, labels)):\n plt.subplot(3, 3, ind + 1)\n image = cv2.imread(os.path.join(BASE_DIR, \"train_images\", image_id))\n image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n\n plt.imshow(image)\n plt.title(f\"Class: {label}\", fontsize=12)\n plt.axis(\"off\")\n \n plt.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Let's visualize a sample of images"},{"metadata":{},"cell_type":"markdown","source":"\n

Data Visualization

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

0 - Complex

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 0]\nprint(f\"Total train images for class 0: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

1 - frog_eye_leaf_spot

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 1]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

2 - frog_eye_leaf_spot complex

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 2]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

3 - healthy

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 3]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

4 - powdery_mildew

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 4]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

5 - powdery_mildew complex

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 5]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

6 - rust

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 6]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

7 - rust complex

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 7]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

8 - rust frog_eye_leaf_spot

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 8]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

9 - scab

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 9]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

10 - scab frog_eye_leaf_spot

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 10]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n

11 - scab frog_eye_leaf_spot complex

"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"tmp_df = train_df[train_df[\"labels_code\"] == 11]\nprint(f\"Total train images for class 1: {tmp_df.shape[0]}\")\n\ntmp_df = tmp_df.sample(9)\nimage_ids = tmp_df[\"image\"].values\nlabels = tmp_df[\"labels\"].values\n\nvisualize_batch(image_ids, labels)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Next steps:\n1. Data augmentation using [albumentations](https://albumentations.ai/) to address class imbalance problem.\n2. Study related literature to aquire domain knowledge.\n3. Implement a baseline model to get an idea.\n4. Create submission method.\n\nFeel free to fork and edit the notebook.\n\nspecial thanks to [Yaroslav Isaienkov](https://www.kaggle.com/ihelon/cassava-leaf-disease-exploratory-data-analysis)\n\nWork in progress"}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":4}