{"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_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"## Happywhale - Whale and Dolphin Identification: Identify whales and dolphins by unique characteristics\n![](https://storage.googleapis.com/kaggle-competitions/kaggle/22962/logos/header.png?t=2021-03-17-22-44-09\")","metadata":{"execution":{"iopub.status.busy":"2022-03-19T13:49:59.693770Z","iopub.execute_input":"2022-03-19T13:49:59.694126Z","iopub.status.idle":"2022-03-19T13:50:00.462567Z","shell.execute_reply.started":"2022-03-19T13:49:59.694037Z","shell.execute_reply":"2022-03-19T13:50:00.461725Z"}}},{"cell_type":"markdown","source":"## Description\nWe use fingerprints and facial recognition to identify people, but can we use similar approaches with animals? In fact, researchers manually track marine life by the shape and markings on their tails, dorsal fins, heads and other body parts. Identification by natural markings via photographs—known as photo-ID—is a powerful tool for marine mammal science. It allows individual animals to be tracked over time and enables assessments of population status and trends. With your help to automate whale and dolphin photo-ID, researchers can reduce image identification times by over 99%. More efficient identification could enable a scale of study previously unaffordable or impossible.\n\nCurrently, most research institutions rely on time-intensive—and sometimes inaccurate—manual matching by the human eye. Thousands of hours go into manual matching, which involves staring at photos to compare one individual to another, finding matches, and identifying new individuals. While researchers enjoy looking at a whale photo or two, manual matching limits the scope and reach.\n\nAlgorithms developed in this competition will be implemented in Happywhale, a research collaboration and citizen science web platform. Its mission is to increase global understanding and caring for marine environments through high quality conservation science and education. Happywhale aims to make it easy and rewarding for the public to participate in science by building innovative tools to engage anyone interested in marine mammals. The platform also serves the research community with powerful collaborative tools.\n\nIn this competition, you’ll develop a model to match individual whales and dolphins by unique—but often subtle—characteristics of their natural markings. You'll pay particular attention to dorsal fins and lateral body views in image sets from a multi-species dataset built by 28 research institutions. The best submissions will suggest photo-ID solutions that are fast and accurate.\n\nIf successful, you'll have a hand in building advanced technology to better understand and manage the impact on the world’s changing oceans. Previous automation attempts resulted in a global database of over 50,000 whales and an agreement with cruise ships to operate at a maximum speed of 11 mph in the most whale-rich region. Your ideas to automate the identification of marine life will help overcome increasing human impacts on oceans, providing a critical tool for conservation science. If there's a whale, there's a way!","metadata":{}},{"cell_type":"markdown","source":"## Data Description\nIn the previous HappyWhale competition, the task was to predict individual humpback whales from images of their flukes. Whales and dolphins in this dataset can be identified by shapes, features and markings (some natural, some acquired) of dorsal fins, backs, heads and flanks. Some species and some individuals have highly distinct features, others are very much less distinct. Further, individual features may change over time. This competition expands that task significantly: data in this competition contains images of over 15,000 unique individual marine mammals from 30 different species collected from 28 different research organizations. Individuals have been manually identified and given an individual_id by marine researches, and your task is to correctly identify these individuals in images. It's a challenging task that has the potential to drive significant advancements in understanding and protecting marine mammals across the globe.\n\nAn important note about data quality: Bringing together this dataset from many different research organization posed a number of practical challenges. Significant effort has been made to minimize data quality issues and as well as to minimize leakage as much as possible. There are undoubtably issues. We encourage the community to report these things so that future versions of the data can be improved, but unless there is a significant issue, we don't expect to make updates to the data during the competition.\n\n## Files\n>- train_images/ - a folder containing the training images\n>- train.csv - provides the species and the individual_id for each of the training images\n>- test_images/ - a folder containing the test images; for each image, your task is to predict the individual_id; no species information is given for the test data; there are individuals in the test data that are not observed in the training data, which should be predicted as new_individual.\nsample_submission.csv - a sample submission file in the correct format\n\n## Evaluation metric\n![image.png](attachment:c4602a14-3335-4d4d-844d-c78faf8a2e10.png)","metadata":{},"attachments":{"c4602a14-3335-4d4d-844d-c78faf8a2e10.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"## Import packages","metadata":{}},{"cell_type":"code","source":"# Install packages","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:02.625768Z","iopub.execute_input":"2022-03-20T21:12:02.626147Z","iopub.status.idle":"2022-03-20T21:12:02.653006Z","shell.execute_reply.started":"2022-03-20T21:12:02.626042Z","shell.execute_reply":"2022-03-20T21:12:02.651693Z"},"trusted":true},"execution_count":null,"outputs":[]},{"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 os\nfrom matplotlib_venn import venn2, venn2_circles, venn2_unweighted\nfrom matplotlib_venn import venn3, venn3_circles\nimport plotly.express as px\nfrom plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot\nimport plotly as py\nimport plotly.graph_objs as go\ninit_notebook_mode(connected=True)\nfrom matplotlib import pyplot as plt\nfrom plotly.subplots import make_subplots\nimport torchvision\nimport torchvision.datasets as dset\nimport torchvision.transforms as transforms\nfrom torch.utils.data import DataLoader,Dataset\nimport matplotlib.pyplot as plt\nimport torchvision.utils \nfrom PIL import Image\nimport torch\nfrom torch.autograd import Variable\nimport PIL.ImageOps    \nimport torch.nn as nn\nfrom torch import optim\nimport torch.nn.functional as F\nimport cv2","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:02.654837Z","iopub.execute_input":"2022-03-20T21:12:02.655174Z","iopub.status.idle":"2022-03-20T21:12:07.889402Z","shell.execute_reply.started":"2022-03-20T21:12:02.655137Z","shell.execute_reply":"2022-03-20T21:12:07.888605Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Load data","metadata":{}},{"cell_type":"code","source":"train_data = pd.read_csv('../input/happy-whale-and-dolphin/train.csv')\nsample_submission_data = pd.read_csv('../input/happy-whale-and-dolphin/sample_submission.csv')","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:07.890672Z","iopub.execute_input":"2022-03-20T21:12:07.891342Z","iopub.status.idle":"2022-03-20T21:12:08.049646Z","shell.execute_reply.started":"2022-03-20T21:12:07.891299Z","shell.execute_reply":"2022-03-20T21:12:08.048828Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## EDA","metadata":{}},{"cell_type":"code","source":"print(train_data.shape, train_data['image'].nunique())\ntrain_data.head()","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:08.052810Z","iopub.execute_input":"2022-03-20T21:12:08.053019Z","iopub.status.idle":"2022-03-20T21:12:08.089033Z","shell.execute_reply.started":"2022-03-20T21:12:08.052992Z","shell.execute_reply":"2022-03-20T21:12:08.088358Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(sample_submission_data.shape, sample_submission_data['image'].nunique())\ndisplay(sample_submission_data['predictions'].iloc[0])\nsample_submission_data.head()","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:08.091425Z","iopub.execute_input":"2022-03-20T21:12:08.091749Z","iopub.status.idle":"2022-03-20T21:12:08.112098Z","shell.execute_reply.started":"2022-03-20T21:12:08.091710Z","shell.execute_reply":"2022-03-20T21:12:08.111266Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Overlap of IDs - train and test datasets","metadata":{}},{"cell_type":"code","source":"set_numbers_train = set(train_data['image'].tolist())\nset_numbers_test = set(sample_submission_data['image'].tolist())\nvenn2((set_numbers_train, set_numbers_test), set_labels = ('Train IDs', 'Test IDs'))","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:08.113436Z","iopub.execute_input":"2022-03-20T21:12:08.113705Z","iopub.status.idle":"2022-03-20T21:12:08.240801Z","shell.execute_reply.started":"2022-03-20T21:12:08.113671Z","shell.execute_reply":"2022-03-20T21:12:08.238873Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Sample images","metadata":{}},{"cell_type":"code","source":"train_data['species'] = np.where(train_data['species'].isin(['beluga','globis']), train_data['species'] + '_whale', train_data['species'])\ntrain_data['Dolphin or Whale'] = train_data['species'].apply(lambda x: 'Dolphin' if x.split('_')[-1]=='dolphin' else 'Whale')\n\ntemp = train_data['species'].value_counts().reset_index()\ntemp['index'] = np.where(temp['index'].isin(['beluga','globis']), temp['index'] + '_whale', temp['index'])\ntemp['Dolphin or Whale'] = temp['index'].apply(lambda x: 'Dolphin' if x.split('_')[-1]=='dolphin' else 'Whale')\ntemp.groupby(['Dolphin or Whale']).agg({'index':'nunique'}).reset_index().rename(columns = {'index':'# Distinct species'})","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:08.242191Z","iopub.execute_input":"2022-03-20T21:12:08.242441Z","iopub.status.idle":"2022-03-20T21:12:08.341705Z","shell.execute_reply.started":"2022-03-20T21:12:08.242405Z","shell.execute_reply":"2022-03-20T21:12:08.340941Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print('\\nSample dolphin images\\n')\nimage_ids = train_data[train_data['Dolphin or Whale']=='Dolphin'].groupby(\"species\").sample(n=1, random_state=1)\n    \nrows = 2\ncols = 5\naxes=[]\nfig=plt.figure(figsize=(50, 25))\n\ncount = 0\nfor a in range(rows*cols):\n    img_path = '../input/happy-whale-and-dolphin/train_images//' + train_data[train_data['image']==(image_ids['image'].iloc[count])]['image'].iloc[0]\n    axes.append(fig.add_subplot(rows, cols, a+1))\n    axes[-1].set_title(image_ids['species'].iloc[count], fontsize=30)\n    plt.imshow(plt.imread(img_path))\n    count +=1\n    \nfig.tight_layout()    \nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:08.343214Z","iopub.execute_input":"2022-03-20T21:12:08.343709Z","iopub.status.idle":"2022-03-20T21:12:17.418403Z","shell.execute_reply.started":"2022-03-20T21:12:08.343667Z","shell.execute_reply":"2022-03-20T21:12:17.416435Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print('\\nSample whale images\\n')\nimage_ids = train_data[train_data['Dolphin or Whale']=='Whale'].groupby(\"species\").sample(n=1, random_state=1)\n    \nrows = 4\ncols = 5\naxes=[]\nfig=plt.figure(figsize=(50, 25))\n\ncount = 0\nfor a in range(rows*cols):\n    img_path = '../input/happy-whale-and-dolphin/train_images//' + train_data[train_data['image']==(image_ids['image'].iloc[count])]['image'].iloc[0]\n    axes.append(fig.add_subplot(rows, cols, a+1))\n    axes[-1].set_title(image_ids['species'].iloc[count], fontsize=30)\n    plt.imshow(plt.imread(img_path))\n    count +=1\n    \nfig.tight_layout()    \nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:17.419735Z","iopub.execute_input":"2022-03-20T21:12:17.420066Z","iopub.status.idle":"2022-03-20T21:12:38.435942Z","shell.execute_reply.started":"2022-03-20T21:12:17.420001Z","shell.execute_reply":"2022-03-20T21:12:38.425989Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Modelling","metadata":{}},{"cell_type":"code","source":"Config = {'train_number_epochs': 100,\n         'batch_size': 8}","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:38.437186Z","iopub.execute_input":"2022-03-20T21:12:38.437875Z","iopub.status.idle":"2022-03-20T21:12:38.441935Z","shell.execute_reply.started":"2022-03-20T21:12:38.437831Z","shell.execute_reply":"2022-03-20T21:12:38.441329Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class SiameseNetwork(torch.nn.Module):\n    def __init__(self):\n        super(SiameseNetwork, self).__init__()\n        self.cnn1 = nn.Sequential(nn.Conv2d(1, 64, 3),\n                                  nn.Conv2d(64, 64, 3),\n                                  nn.Conv2d(64, 128, 3),\n                                  nn.Conv2d(128, 128, 3),\n                                  nn.Conv2d(128, 256, 3),\n                                  nn.Conv2d(256, 256, 3),\n                                  nn.MaxPool2d(2, 2),\n                                  nn.Flatten(),\n                                  nn.Linear(3748096, 4096)).cuda()\n\n    def forward_once(self, x):\n        output = self.cnn1(x.float())\n        return output\n\n    def forward(self, input1, input2):\n        output1 = self.forward_once(input1)\n        output2 = self.forward_once(input2)\n        return output1, output2","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:38.442965Z","iopub.execute_input":"2022-03-20T21:12:38.443498Z","iopub.status.idle":"2022-03-20T21:12:38.454424Z","shell.execute_reply.started":"2022-03-20T21:12:38.443458Z","shell.execute_reply":"2022-03-20T21:12:38.453573Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class ContrastiveLoss(torch.nn.Module):\n    \"\"\"\n    Contrastive loss function.\n    Based on: http://yann.lecun.com/exdb/publis/pdf/hadsell-chopra-lecun-06.pdf\n    \"\"\"\n\n    def __init__(self, margin=2.0):\n        super(ContrastiveLoss, self).__init__()\n        self.margin = margin\n\n    def forward(self, output1, output2, label):\n        euclidean_distance = F.pairwise_distance(output1, output2)\n        loss_contrastive = torch.mean((1-label) * torch.pow(euclidean_distance, 2) +\n                                      (label) * torch.pow(torch.clamp(self.margin - euclidean_distance, min=0.0), 2))\n\n\n        return loss_contrastive","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:38.455988Z","iopub.execute_input":"2022-03-20T21:12:38.456232Z","iopub.status.idle":"2022-03-20T21:12:38.468831Z","shell.execute_reply.started":"2022-03-20T21:12:38.456201Z","shell.execute_reply":"2022-03-20T21:12:38.464513Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"img_shape = (256, 256)  # The image shape used by the model\nclass SiameseNetworkDataset():    \n    def __init__(self,imageFolderDataset,transform=None,should_invert=True):\n        self.images_path = f'../input/happy-whale-and-dolphin/{imageFolderDataset}_images'\n        self.imageFolderDataset = os.listdir(self.images_path)    \n        self.transform = transform\n        self.should_invert = should_invert\n        \n    def __getitem__(self,index):\n        img0_tuple = self.imageFolderDataset[index]\n        img1_tuple = np.random.choice(self.imageFolderDataset)\n\n        img0 = plt.imread(self.images_path + '//' + img0_tuple)\n        img1 = plt.imread(self.images_path + '//' + img1_tuple)\n        \n        if len(img0.shape)==3:\n            img0 = (cv2.resize(img0, img_shape) / 255)\n        else:\n            img0 = cv2.merge((img0, img0, img0))\n            img0 = (cv2.resize(img0, img_shape) / 255)\n            \n        if len(img1.shape)==3:\n            img1 = (cv2.resize(img1, img_shape) / 255)\n        else:\n            img1 = cv2.merge((img1, img1, img1))\n            img1 = (cv2.resize(img1, img_shape) / 255)\n            \n        img0 = np.rollaxis(img0, 2, 0)[0:1,:,:]   \n        img1 = np.rollaxis(img1, 2, 0)[0:1,:,:]\n            \n        if self.transform is not None:\n            img0 = self.transform(img0)\n            img1 = self.transform(img1)\n        \n        return torch.tensor(img0), torch.tensor(img1), torch.from_numpy(np.array([int(img1_tuple[1]!=img0_tuple[1])],dtype=np.float32))\n    \n    def __len__(self):\n        return len(self.imageFolderDataset)","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:38.470589Z","iopub.execute_input":"2022-03-20T21:12:38.470835Z","iopub.status.idle":"2022-03-20T21:12:38.488263Z","shell.execute_reply.started":"2022-03-20T21:12:38.470804Z","shell.execute_reply":"2022-03-20T21:12:38.487240Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"siamese_dataset = SiameseNetworkDataset(imageFolderDataset='train',\n                                        transform=None,\n                                        should_invert=False)\n\ntrain_dataloader = DataLoader(siamese_dataset,\n                        shuffle=True,\n                        num_workers=8,\n                        batch_size=Config['batch_size'])","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:38.491803Z","iopub.execute_input":"2022-03-20T21:12:38.492113Z","iopub.status.idle":"2022-03-20T21:12:39.186779Z","shell.execute_reply.started":"2022-03-20T21:12:38.492077Z","shell.execute_reply":"2022-03-20T21:12:39.185935Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def show_plot(iteration,loss):\n    plt.plot(iteration,loss)\n    plt.show()","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:39.188227Z","iopub.execute_input":"2022-03-20T21:12:39.189046Z","iopub.status.idle":"2022-03-20T21:12:39.193737Z","shell.execute_reply.started":"2022-03-20T21:12:39.188993Z","shell.execute_reply":"2022-03-20T21:12:39.192968Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"net = SiameseNetwork().cuda()\ncriterion = ContrastiveLoss()\noptimizer = optim.Adam(net.parameters(),lr = 0.0005)\n\ncounter = []\nloss_history = [] \niteration_number= 0\n\nfor epoch in range(0,Config['train_number_epochs']):\n    for i, data in enumerate(train_dataloader,0):\n        img0, img1 , label = data\n        img0, img1 , label = Variable(img0).cuda(), Variable(img1).cuda() , Variable(label).cuda()\n        output1,output2 = net(img0,img1)\n        optimizer.zero_grad()\n        loss_contrastive = criterion(output1,output2,label)\n        loss_contrastive.backward()\n        optimizer.step()\n        if i %10 == 0 :\n            print(\"Epoch number {}\\n Current loss {}\\n\".format(epoch,loss_contrastive.data[0]))\n            iteration_number +=10\n            counter.append(iteration_number)\n            loss_history.append(loss_contrastive.data[0])\nshow_plot(counter,loss_history)","metadata":{"execution":{"iopub.status.busy":"2022-03-20T21:12:39.195060Z","iopub.execute_input":"2022-03-20T21:12:39.195754Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"folder_dataset_test = dset.ImageFolder(root=Config.testing_dir)\nsiamese_dataset = SiameseNetworkDataset(imageFolderDataset=folder_dataset_test,\n                                        transform=transforms.Compose([transforms.Scale((100,100)),\n                                                                      transforms.ToTensor()\n                                                                      ])\n                                       ,should_invert=False)\n\ntest_dataloader = DataLoader(siamese_dataset,num_workers=6,batch_size=1,shuffle=True)\ndataiter = iter(test_dataloader)\nx0,_,_ = next(dataiter)\n\nfor i in range(10):\n    _,x1,label2 = next(dataiter)\n    concatenated = torch.cat((x0,x1),0)\n    \n    output1,output2 = net(Variable(x0).cuda(),Variable(x1).cuda())\n    euclidean_distance = F.pairwise_distance(output1, output2)\n    imshow(torchvision.utils.make_grid(concatenated),'Dissimilarity: {:.2f}'.format(euclidean_distance.cpu().data.numpy()[0][0]))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}