1.1 | THE NEED
\n1.2 | PROBLEM STATEMENT
\n1.3 | STUDY AIM
\n1.4 | PLAYGROUND PREDICTION COMPETITION
\n0b2ea2a822ad23fdb1b5dd26653da899fbd2c0d5,0","metadata":{}},{"cell_type":"code","source":"!pip install torchsummary","metadata":{"_kg_hide-output":true,"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-03-04T20:46:29.330654Z","iopub.execute_input":"2022-03-04T20:46:29.330922Z","iopub.status.idle":"2022-03-04T20:46:38.633544Z","shell.execute_reply.started":"2022-03-04T20:46:29.330889Z","shell.execute_reply":"2022-03-04T20:46:38.632739Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"We have a table for identification, id & subsequent expert label diagnosis label","metadata":{}},{"cell_type":"code","source":"labels_df = pd.read_csv('/kaggle/input/histopathologic-cancer-detection/train_labels.csv')\nlabels_df.head()","metadata":{"execution":{"iopub.status.busy":"2022-03-04T20:46:38.635808Z","iopub.execute_input":"2022-03-04T20:46:38.63609Z","iopub.status.idle":"2022-03-04T20:46:39.230978Z","shell.execute_reply.started":"2022-03-04T20:46:38.636052Z","shell.execute_reply":"2022-03-04T20:46:39.230277Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# No duplicate ids found\nlabels_df[labels_df.duplicated(keep=False)]","metadata":{"execution":{"iopub.status.busy":"2022-03-04T20:46:39.232738Z","iopub.execute_input":"2022-03-04T20:46:39.232992Z","iopub.status.idle":"2022-03-04T20:46:39.309554Z","shell.execute_reply.started":"2022-03-04T20:46:39.232958Z","shell.execute_reply":"2022-03-04T20:46:39.308732Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Definitely not as one sides as was expected, but the dataset still favours non malignant, normal cases (13k) compared to malignant cases (8.9k)","metadata":{}},{"cell_type":"code","source":"labels_df['label'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2022-03-04T20:46:39.310866Z","iopub.execute_input":"2022-03-04T20:46:39.31118Z","iopub.status.idle":"2022-03-04T20:46:39.320952Z","shell.execute_reply.started":"2022-03-04T20:46:39.311144Z","shell.execute_reply":"2022-03-04T20:46:39.320056Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"1.5 | DATA SAMPLES
\nDataset class by subclassing the __Pytorch Dataset__ class.\n- We need two essential fuctions __len__ & __getitem__ in our custom class.","metadata":{}},{"cell_type":"code","source":"# fix torch random seed\ntorch.manual_seed(0)\n\nclass pytorch_data(Dataset):\n \n def __init__(self, data_dir, transform,data_type=\"train\"): \n \n # Get Image File Names\n cdm_data=os.path.join(data_dir,data_type) # directory of files\n file_names = os.listdir(cdm_data) # get list of images in that directory\n self.full_filenames = [os.path.join(cdm_data, f) for f in file_names] # get the full path to images\n \n # Get Labels\n labels_data=os.path.join(data_dir,\"train_labels.csv\") \n labels_df=pd.read_csv(labels_data)\n labels_df.set_index(\"id\", inplace=True) # set data frame index to id\n # obtained labels from df\n self.labels = [labels_df.loc[filename[:-4]].values[0] for filename in file_names] \n self.transform = transform\n \n def __len__(self):\n return len(self.full_filenames) # size of dataset\n \n def __getitem__(self, idx):\n # open image, apply transforms and return with label\n image = Image.open(self.full_filenames[idx]) # Open Image with PIL\n image = self.transform(image) # Apply Specific Transformation to Image\n return image, self.labels[idx]","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:36:58.245452Z","iopub.execute_input":"2022-03-03T19:36:58.245837Z","iopub.status.idle":"2022-03-03T19:36:58.260948Z","shell.execute_reply.started":"2022-03-03T19:36:58.245801Z","shell.execute_reply":"2022-03-03T19:36:58.260021Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# define transformation that converts a PIL image into PyTorch tensors.\nimport torchvision.transforms as transforms\ndata_transformer = transforms.Compose([transforms.ToTensor(),\n transforms.Resize((46,46))])","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:36:58.262695Z","iopub.execute_input":"2022-03-03T19:36:58.26323Z","iopub.status.idle":"2022-03-03T19:36:58.269901Z","shell.execute_reply.started":"2022-03-03T19:36:58.263186Z","shell.execute_reply":"2022-03-03T19:36:58.269027Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Define an object of the custom dataset for the train folder.\ndata_dir = '/kaggle/input/histopathologic-cancer-detection/'\nimg_dataset = pytorch_data(data_dir, data_transformer, \"train\") # Histopathalogic images","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:36:58.273508Z","iopub.execute_input":"2022-03-03T19:36:58.273953Z","iopub.status.idle":"2022-03-03T19:37:22.713813Z","shell.execute_reply.started":"2022-03-03T19:36:58.273918Z","shell.execute_reply":"2022-03-03T19:37:22.713057Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# load an example tensor\nimg,label=img_dataset[10]\nprint(img.shape,torch.min(img),torch.max(img))","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:22.715091Z","iopub.execute_input":"2022-03-03T19:37:22.715333Z","iopub.status.idle":"2022-03-03T19:37:22.796103Z","shell.execute_reply.started":"2022-03-03T19:37:22.715302Z","shell.execute_reply":"2022-03-03T19:37:22.795341Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 3 | SPLITTING THE DATASET\n\n- Among the training set, we need to evaluate the model on validation datasets to track the model's performance during training.\n- Let's use 20% of img_dataset as the validation dataset & use the rest as the training set.","metadata":{}},{"cell_type":"code","source":"len_img=len(img_dataset)\nlen_train=int(0.8*len_img)\nlen_val=len_img-len_train\n\n# Split Pytorch tensor\ntrain_ts,val_ts=random_split(img_dataset,\n [len_train,len_val]) # random split 80/20\n\nprint(\"train dataset size:\", len(train_ts))\nprint(\"validation dataset size:\", len(val_ts))","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:22.797144Z","iopub.execute_input":"2022-03-03T19:37:22.797377Z","iopub.status.idle":"2022-03-03T19:37:22.825009Z","shell.execute_reply.started":"2022-03-03T19:37:22.797345Z","shell.execute_reply":"2022-03-03T19:37:22.824317Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# getting the torch tensor image & target variable\nii=-1\nfor x,y in train_ts:\n print(x.shape,y)\n ii+=1\n if(ii>5):\n break","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:22.826211Z","iopub.execute_input":"2022-03-03T19:37:22.826441Z","iopub.status.idle":"2022-03-03T19:37:22.928525Z","shell.execute_reply.started":"2022-03-03T19:37:22.826409Z","shell.execute_reply":"2022-03-03T19:37:22.927856Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import plotly.express as px\n\ndef plot_img(x,y):\n\n npimg = x.numpy() # convert tensor to numpy array\n npimg_tr=np.transpose(npimg, (1,2,0)) # Convert to H*W*C shape\n fig = px.imshow(npimg_tr)\n fig.update_layout(coloraxis_showscale=False,title=str(y_grid_train))\n fig.update_xaxes(showticklabels=False)\n fig.update_layout(template='plotly_white',height=200);fig.update_layout(margin={\"r\":0,\"t\":60,\"l\":0,\"b\":0})\n fig.update_layout(title={'text': str(y),'y':0.9,'x':0.5,'xanchor': 'center','yanchor': 'top'})\n \n fig.show()","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-03-03T19:37:22.929714Z","iopub.execute_input":"2022-03-03T19:37:22.930212Z","iopub.status.idle":"2022-03-03T19:37:26.211913Z","shell.execute_reply.started":"2022-03-03T19:37:22.930175Z","shell.execute_reply":"2022-03-03T19:37:26.211153Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Training Subset Examples \nSome examples from our training data subset, with corresponding labels.","metadata":{}},{"cell_type":"code","source":"import plotly.express as px\n\n# Create grid of sample images \ngrid_size=8\nrnd_inds=np.random.randint(0,len(train_ts),grid_size)\nprint(\"image indices:\",rnd_inds)\n\nx_grid_train=[train_ts[i][0] for i in rnd_inds]\ny_grid_train=[train_ts[i][1] for i in rnd_inds]\n\nx_grid_train=utils.make_grid(x_grid_train, nrow=4, padding=2)\nprint(x_grid_train.shape)\n \nplot_img(x_grid_train,y_grid_train)","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:26.213141Z","iopub.execute_input":"2022-03-03T19:37:26.2134Z","iopub.status.idle":"2022-03-03T19:37:27.109123Z","shell.execute_reply.started":"2022-03-03T19:37:26.213367Z","shell.execute_reply":"2022-03-03T19:37:27.108434Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Validation Dataset Examples\nSome examples from the validation subset, with corresponding labels.","metadata":{}},{"cell_type":"code","source":"grid_size=8\nrnd_inds=np.random.randint(0,len(val_ts),grid_size)\nprint(\"image indices:\",rnd_inds)\nx_grid_val=[val_ts[i][0] for i in range(grid_size)]\ny_grid_val=[val_ts[i][1] for i in range(grid_size)]\n\nx_grid_val=utils.make_grid(x_grid_val, nrow=4, padding=2)\nprint(x_grid_val.shape)\n\nplot_img(x_grid_val,y_grid_val)","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:27.110312Z","iopub.execute_input":"2022-03-03T19:37:27.110634Z","iopub.status.idle":"2022-03-03T19:37:27.300969Z","shell.execute_reply.started":"2022-03-03T19:37:27.110596Z","shell.execute_reply":"2022-03-03T19:37:27.29863Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 4 | TRANSFORMING THE DATA\n\n#### **IMAGE AUGMENTATIONS**\n\n- Among with pretrained models, image __transformation__ and __image augmentation__ are generally considered to be an essential parts of constructing deep learning models.\n- Using image transformations, we can expand our dataset or resize and normalise it to achieve better model performance.\n- Typical transformations include __horizontal__,__vertical flipping__, __rotation__, __resizing__.\n- We can use various image transformations for our binary classification model without making label changes; we can flip/rotate a __malignant__ image but it will remain the same, __malignant__.\n- We can use the torchvision module to perform image transformations during the training process.\n\n#### **TRAINING DATA AUGMENTATIONS**\n- transforms.RandomHorizontalFlip(p=0.5): Flips the image horizontally with the probability of 0.5\n- transforms.RandomVerticalFlip(p=0.5) : Flips the image vertically \" \n- transforms.RandomRotation(45) : Rotates the images in the range of (-45,45) degrees.\n- transforms.RandomResizedCrop(96,scale=(0.8,1.0),ratio=(1.0,1.0)) : Randomly square crops the image in the range of [72,96], followed by a resize to 96x96, which is the original pixel size of our image data.\n- transforms.ToTensor() : Converts to Tensor & Normalises as shown above already.","metadata":{}},{"cell_type":"code","source":"# Define the following transformations for the training dataset\ntr_transf = transforms.Compose([\n# transforms.Resize((40,40)),\n transforms.RandomHorizontalFlip(p=0.5), \n transforms.RandomVerticalFlip(p=0.5), \n transforms.RandomRotation(45), \n# transforms.RandomResizedCrop(50,scale=(0.8,1.0),ratio=(1.0,1.0)),\n transforms.ToTensor()])","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:27.302275Z","iopub.execute_input":"2022-03-03T19:37:27.302539Z","iopub.status.idle":"2022-03-03T19:37:27.307737Z","shell.execute_reply.started":"2022-03-03T19:37:27.302505Z","shell.execute_reply":"2022-03-03T19:37:27.306787Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# For the validation dataset, we don't need any augmentation; simply convert images into tensors\nval_transf = transforms.Compose([\n transforms.ToTensor()])\n\n# After defining the transformations, overwrite the transform functions of train_ts, val_ts\ntrain_ts.transform=tr_transf\nval_ts.transform=val_transf","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:27.308937Z","iopub.execute_input":"2022-03-03T19:37:27.309688Z","iopub.status.idle":"2022-03-03T19:37:27.31842Z","shell.execute_reply.started":"2022-03-03T19:37:27.309633Z","shell.execute_reply":"2022-03-03T19:37:27.317708Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 5 | CREATING DATALOADERS\n\n- Ready to create a PyTorch Dataloader. If we don't use __Dataloaders__, we have to write code to loop over datasets & extract a data batch; automated.\n- We need to define a __batch_size__ : The number of images extracted from the dataset each iteration","metadata":{}},{"cell_type":"code","source":"from torch.utils.data import DataLoader\ntrain_dl = DataLoader(train_ts, batch_size=32, shuffle=True)\nval_dl = DataLoader(val_ts, batch_size=32, shuffle=False)","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:27.3196Z","iopub.execute_input":"2022-03-03T19:37:27.320012Z","iopub.status.idle":"2022-03-03T19:37:27.327615Z","shell.execute_reply.started":"2022-03-03T19:37:27.319976Z","shell.execute_reply":"2022-03-03T19:37:27.326939Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for x,y in train_dl:\n print(x.shape,y)\n break","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:27.328782Z","iopub.execute_input":"2022-03-03T19:37:27.329152Z","iopub.status.idle":"2022-03-03T19:37:27.764603Z","shell.execute_reply.started":"2022-03-03T19:37:27.329116Z","shell.execute_reply":"2022-03-03T19:37:27.762888Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 6 | CNN BINARY CLASSIFICATION MODEL\n\n- Model is comprised of __four CNN__ & __two fully connected__ layers.\n- After each convolution layer, there is a pooling layer, implementation in PyTorch.\n- Pooling layers are used after each convolution layer as well.","metadata":{}},{"cell_type":"code","source":"# Function to calculate the output size of a CNN layer\n# before making it an input into the linear layer\n\ndef findConv2dOutShape(hin,win,conv,pool=2):\n # get conv arguments\n kernel_size=conv.kernel_size\n stride=conv.stride\n padding=conv.padding\n dilation=conv.dilation\n\n hout=np.floor((hin+2*padding[0]-dilation[0]*(kernel_size[0]-1)-1)/stride[0]+1)\n wout=np.floor((win+2*padding[1]-dilation[1]*(kernel_size[1]-1)-1)/stride[1]+1)\n\n if pool:\n hout/=pool\n wout/=pool\n return int(hout),int(wout)","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-03-03T19:37:27.766122Z","iopub.execute_input":"2022-03-03T19:37:27.766392Z","iopub.status.idle":"2022-03-03T19:37:27.773133Z","shell.execute_reply.started":"2022-03-03T19:37:27.766355Z","shell.execute_reply":"2022-03-03T19:37:27.772147Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import torch.nn as nn\nimport torch.nn.functional as F\n\nclass Network(nn.Module):\n \n # Network Initialisation\n def __init__(self, params):\n \n super(Network, self).__init__()\n \n Cin,Hin,Win=params[\"shape_in\"]\n init_f=params[\"initial_filters\"] \n num_fc1=params[\"num_fc1\"] \n num_classes=params[\"num_classes\"] \n self.dropout_rate=params[\"dropout_rate\"] \n \n # Convolution Layers\n self.conv1 = nn.Conv2d(Cin, init_f, kernel_size=3)\n h,w=findConv2dOutShape(Hin,Win,self.conv1)\n self.conv2 = nn.Conv2d(init_f, 2*init_f, kernel_size=3)\n h,w=findConv2dOutShape(h,w,self.conv2)\n self.conv3 = nn.Conv2d(2*init_f, 4*init_f, kernel_size=3)\n h,w=findConv2dOutShape(h,w,self.conv3)\n self.conv4 = nn.Conv2d(4*init_f, 8*init_f, kernel_size=3)\n h,w=findConv2dOutShape(h,w,self.conv4)\n \n # compute the flatten size\n self.num_flatten=h*w*8*init_f\n self.fc1 = nn.Linear(self.num_flatten, num_fc1)\n self.fc2 = nn.Linear(num_fc1, num_classes)\n\n def forward(self,X):\n \n # Convolution & Pool Layers\n X = F.relu(self.conv1(X)); X = F.max_pool2d(X, 2, 2)\n X = F.relu(self.conv2(X)); X = F.max_pool2d(X, 2, 2)\n X = F.relu(self.conv3(X));X = F.max_pool2d(X, 2, 2)\n X = F.relu(self.conv4(X));X = F.max_pool2d(X, 2, 2)\n\n X = X.view(-1, self.num_flatten)\n \n X = F.relu(self.fc1(X))\n X=F.dropout(X, self.dropout_rate)\n X = self.fc2(X)\n return F.log_softmax(X, dim=1)\n\n# Neural Network Predefined Parameters\nparams_model={\n \"shape_in\": (3,46,46), \n \"initial_filters\": 8, \n \"num_fc1\": 100,\n \"dropout_rate\": 0.25,\n \"num_classes\": 2}\n\n# Create instantiation of Network class\ncnn_model = Network(params_model)\n\n# define computation hardware approach (GPU/CPU)\ndevice = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\nmodel = cnn_model.to(device)","metadata":{"_kg_hide-input":false,"execution":{"iopub.status.busy":"2022-03-03T19:37:27.774599Z","iopub.execute_input":"2022-03-03T19:37:27.774875Z","iopub.status.idle":"2022-03-03T19:37:30.984225Z","shell.execute_reply.started":"2022-03-03T19:37:27.774842Z","shell.execute_reply":"2022-03-03T19:37:30.983465Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 7 | DEFINING A LOSS FUNCTION\n\n- Loss Functions are one of the key pieces of an effective deep learning solution.\n- Pytorch uses loss functions to determine how it will update the network to reach the desired solution.\n- The standard loss function for classification tasks is __cross entropy loss__ or __logloss__\n- When defining a loss function, we need to consider, the number of model outputs and their activation functions.\n- For binary classification tasks, we can choose one or two outputs.\n- It is recommended to use __log_softmax__ as it is easier to expand to multiclass classification; PyTorch combines the log and softmax operations into one function, due to numerical stability and speed.","metadata":{}},{"cell_type":"code","source":"loss_func = nn.NLLLoss(reduction=\"sum\")","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:30.985341Z","iopub.execute_input":"2022-03-03T19:37:30.985593Z","iopub.status.idle":"2022-03-03T19:37:30.990942Z","shell.execute_reply.started":"2022-03-03T19:37:30.985561Z","shell.execute_reply":"2022-03-03T19:37:30.99005Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 8 | DEFINING AN OPTIMISER\n\n- Training the network involves passing data through the network, using the loss function to determine the difference b/w the prediction & true value. Which is then followed by using of that information to update the weights of the network in an attempt to make the loss function return as small of a loss as possible. __Performing updates on the neural network, an optimiser is used__.\n- The torch.optim contains implementations of common optimisers\n- The optimiser will hold the current state and will update the parameters based on the computed gradients\n- For binary classification taskss, __SGD__, __Adam__ Optimisers are commonly used, let's use the latter here.","metadata":{}},{"cell_type":"code","source":"from torch import optim\nopt = optim.Adam(cnn_model.parameters(), lr=3e-4)\nlr_scheduler = ReduceLROnPlateau(opt, mode='min',factor=0.5, patience=20,verbose=1)","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:30.992028Z","iopub.execute_input":"2022-03-03T19:37:30.992719Z","iopub.status.idle":"2022-03-03T19:37:31.003637Z","shell.execute_reply.started":"2022-03-03T19:37:30.992682Z","shell.execute_reply":"2022-03-03T19:37:31.003005Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 9 | TRAINING & EVALUATION","metadata":{}},{"cell_type":"code","source":"''' Helper Functions'''\n\n# Function to get the learning rate\ndef get_lr(opt):\n for param_group in opt.param_groups:\n return param_group['lr']\n\n# Function to compute the loss value per batch of data\ndef loss_batch(loss_func, output, target, opt=None):\n \n loss = loss_func(output, target) # get loss\n pred = output.argmax(dim=1, keepdim=True) # Get Output Class\n metric_b=pred.eq(target.view_as(pred)).sum().item() # get performance metric\n \n if opt is not None:\n opt.zero_grad()\n loss.backward()\n opt.step()\n\n return loss.item(), metric_b\n\n# Compute the loss value & performance metric for the entire dataset (epoch)\ndef loss_epoch(model,loss_func,dataset_dl,check=False,opt=None):\n \n run_loss=0.0 \n t_metric=0.0\n len_data=len(dataset_dl.dataset)\n\n # internal loop over dataset\n for xb, yb in dataset_dl:\n # move batch to device\n xb=xb.to(device)\n yb=yb.to(device)\n output=model(xb) # get model output\n loss_b,metric_b=loss_batch(loss_func, output, yb, opt) # get loss per batch\n run_loss+=loss_b # update running loss\n\n if metric_b is not None: # update running metric\n t_metric+=metric_b\n\n # break the loop in case of sanity check\n if check is True:\n break\n \n loss=run_loss/float(len_data) # average loss value\n metric=t_metric/float(len_data) # average metric value\n \n return loss, metric","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-03-03T19:37:31.004974Z","iopub.execute_input":"2022-03-03T19:37:31.005223Z","iopub.status.idle":"2022-03-03T19:37:31.01716Z","shell.execute_reply.started":"2022-03-03T19:37:31.00519Z","shell.execute_reply":"2022-03-03T19:37:31.016347Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def train_val(model, params,verbose=False):\n \n # Get the parameters\n epochs=params[\"epochs\"]\n loss_func=params[\"f_loss\"]\n opt=params[\"optimiser\"]\n train_dl=params[\"train\"]\n val_dl=params[\"val\"]\n check=params[\"check\"]\n lr_scheduler=params[\"lr_change\"]\n weight_path=params[\"weight_path\"]\n \n loss_history={\"train\": [],\"val\": []} # history of loss values in each epoch\n metric_history={\"train\": [],\"val\": []} # histroy of metric values in each epoch\n best_model_wts = copy.deepcopy(model.state_dict()) # a deep copy of weights for the best performing model\n best_loss=float('inf') # initialize best loss to a large value\n \n # main loop\n for epoch in range(epochs):\n \n ''' Get the Learning Rate '''\n current_lr=get_lr(opt)\n if(verbose):\n print('Epoch {}/{}, current lr={}'.format(epoch, epochs - 1, current_lr))\n \n ''' Train the Model on the Training Set '''\n model.train()\n train_loss, train_metric=loss_epoch(model,loss_func,train_dl,check,opt)\n\n ''' Collect loss and metric for training dataset ''' \n loss_history[\"train\"].append(train_loss)\n metric_history[\"train\"].append(train_metric)\n \n ''' Evaluate model on validation dataset '''\n model.eval()\n with torch.no_grad():\n val_loss, val_metric=loss_epoch(model,loss_func,val_dl,check)\n \n # store best model\n if val_loss < best_loss:\n best_loss = val_loss\n best_model_wts = copy.deepcopy(model.state_dict())\n \n # store weights into a local file\n torch.save(model.state_dict(), weight_path)\n if(verbose):\n print(\"Copied best model weights!\")\n \n # collect loss and metric for validation dataset\n loss_history[\"val\"].append(val_loss)\n metric_history[\"val\"].append(val_metric)\n \n # learning rate schedule\n lr_scheduler.step(val_loss)\n if current_lr != get_lr(opt):\n if(verbose):\n print(\"Loading best model weights!\")\n model.load_state_dict(best_model_wts) \n\n if(verbose):\n print(f\"train loss: {train_loss:.6f}, dev loss: {val_loss:.6f}, accuracy: {100*val_metric:.2f}\")\n print(\"-\"*10) \n\n # load best model weights\n model.load_state_dict(best_model_wts)\n \n return model, loss_history, metric_history","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:31.018671Z","iopub.execute_input":"2022-03-03T19:37:31.01898Z","iopub.status.idle":"2022-03-03T19:37:31.033681Z","shell.execute_reply.started":"2022-03-03T19:37:31.018944Z","shell.execute_reply":"2022-03-03T19:37:31.032897Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Create Parameter Dictionary\nparams_train={\n \"train\": train_dl,\n \"val\": val_dl,\n \"epochs\": 30,\n \"optimiser\": optim.Adam(cnn_model.parameters(),lr=3e-4),\n \"lr_change\": ReduceLROnPlateau(opt,\n mode='min',\n factor=0.5,\n patience=20,verbose=0),\n \"f_loss\": nn.NLLLoss(reduction=\"sum\"),\n \"weight_path\": \"weights.pt\",\n \"check\": True, \n}\n\n# train and validate the model w/ Sanity Check\ncnn_model,loss_hist,metric_hist=train_val(cnn_model,params_train)","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:31.035164Z","iopub.execute_input":"2022-03-03T19:37:31.035426Z","iopub.status.idle":"2022-03-03T19:37:50.861931Z","shell.execute_reply.started":"2022-03-03T19:37:31.035393Z","shell.execute_reply":"2022-03-03T19:37:50.86114Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Train-Validation Progress\nepochs=params_train[\"epochs\"]\n\nfig = make_subplots(rows=1, cols=2,subplot_titles=['lost_hist','metric_hist'])\nfig.add_trace(go.Scatter(x=[*range(1,epochs+1)], y=loss_hist[\"train\"],name='loss_hist[\"train\"]'),row=1, col=1)\nfig.add_trace(go.Scatter(x=[*range(1,epochs+1)], y=loss_hist[\"val\"],name='loss_hist[\"val\"]'),row=1, col=1)\nfig.add_trace(go.Scatter(x=[*range(1,epochs+1)], y=metric_hist[\"train\"],name='metric_hist[\"train\"]'),row=1, col=2)\nfig.add_trace(go.Scatter(x=[*range(1,epochs+1)], y=metric_hist[\"val\"],name='metric_hist[\"val\"]'),row=1, col=2)\nfig.update_layout(template='plotly_white');fig.update_layout(margin={\"r\":0,\"t\":60,\"l\":0,\"b\":0},height=300)\nfig.show()","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-03-03T19:37:50.865959Z","iopub.execute_input":"2022-03-03T19:37:50.866273Z","iopub.status.idle":"2022-03-03T19:37:50.939087Z","shell.execute_reply.started":"2022-03-03T19:37:50.866234Z","shell.execute_reply":"2022-03-03T19:37:50.93844Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"params_train={\n \"train\": train_dl,\"val\": val_dl,\n \"epochs\": 50,\n \"optimiser\": optim.Adam(cnn_model.parameters(),\n lr=3e-4),\n \"lr_change\": ReduceLROnPlateau(opt,\n mode='min',\n factor=0.5,\n patience=20,\n verbose=0),\n \"f_loss\": nn.NLLLoss(reduction=\"sum\"),\n \"weight_path\": \"weights.pt\",\n \"check\": False, \n}\n\n''' Actual Train / Evaluation of CNN Model '''\n# train and validate the model\ncnn_model,loss_hist,metric_hist=train_val(cnn_model,params_train)","metadata":{"execution":{"iopub.status.busy":"2022-03-03T19:37:50.945376Z","iopub.execute_input":"2022-03-03T19:37:50.945625Z","iopub.status.idle":"2022-03-03T19:38:15.056698Z","shell.execute_reply.started":"2022-03-03T19:37:50.945591Z","shell.execute_reply":"2022-03-03T19:38:15.055879Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Train-Validation Progress\nepochs=params_train[\"epochs\"]\n\nfig = make_subplots(rows=1, cols=2,subplot_titles=['lost_hist','metric_hist'])\nfig.add_trace(go.Scatter(x=[*range(1,epochs+1)], y=loss_hist[\"train\"],name='loss_hist[\"train\"]'),row=1, col=1)\nfig.add_trace(go.Scatter(x=[*range(1,epochs+1)], y=loss_hist[\"val\"],name='loss_hist[\"val\"]'),row=1, col=1)\nfig.add_trace(go.Scatter(x=[*range(1,epochs+1)], y=metric_hist[\"train\"],name='metric_hist[\"train\"]'),row=1, col=2)\nfig.add_trace(go.Scatter(x=[*range(1,epochs+1)], y=metric_hist[\"val\"],name='metric_hist[\"val\"]'),row=1, col=2)\nfig.update_layout(template='plotly_white');fig.update_layout(margin={\"r\":0,\"t\":60,\"l\":0,\"b\":0},height=300)\nfig.show()","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2022-03-03T19:38:15.057958Z","iopub.execute_input":"2022-03-03T19:38:15.058207Z","iopub.status.idle":"2022-03-03T19:38:15.118161Z","shell.execute_reply.started":"2022-03-03T19:38:15.058175Z","shell.execute_reply":"2022-03-03T19:38:15.117516Z"},"trusted":true},"execution_count":null,"outputs":[]}]}