Adding new py files and source data

Author: t129047

PyTorch_1_Datasets.py

@@ -0,0 +1,33 @@
+import torch
+from torch.utils.data import Dataset
+
+#crate a subclass from Dataset
+class toy_set(Dataset):
+    #contructor
+    def __init__(self,length = 50,transform = None):
+        self.len = length
+        # pylint: disable=E1101 
+        self.x = torch.ones(length,2)
+        self.y = torch.ones(length,1)
+        # pylint: enable=E1101
+        self.transform = transform
+
+    #return data at a given index
+    def __getitem__(self,index):
+        sample = self.x[index],self.y[index]
+        if self.transform:
+            sample = self.transform(sample)
+        return sample
+
+    #return length
+    def __len__(self):
+        return self.len
+
+our_dataset = toy_set()
+
+for idx in range(3):
+    x,y = our_dataset[idx]
+    print("x = {} y = {}".format(x,y))
+
+for x,y in our_dataset:
+    print("x = {} y = {}".format(x,y))

PyTorch_2_Transforms.py

@@ -0,0 +1,55 @@
+import torch
+from torch.utils.data import Dataset
+
+#crate a subclass from Dataset
+class toy_set(Dataset):
+    #contructor
+    def __init__(self,length = 50,transform = None):
+        self.len = length
+        # pylint: disable=E1101 
+        self.x = torch.ones(length,2)
+        self.y = torch.ones(length,1)
+        # pylint: enable=E1101
+        self.transform = transform
+
+    #return data at a given index
+    def __getitem__(self,index):
+        sample = self.x[index],self.y[index]
+        if self.transform:
+            sample = self.transform(sample)
+        return sample
+
+    #return length
+    def __len__(self):
+        return self.len
+
+#crate a transforms class
+class add_mult(object):
+    #contructor
+    def __init__(self,addx = 1,muly = 2):
+        self.addx = addx
+        self.muly = muly
+    #executor
+    def __call__(self,sample):
+        return sample[0] + self.addx , sample[1] * self.muly
+
+#create add_mult transforms object and toy_set Dataset object
+a_m = add_mult()
+data_set = toy_set()
+
+#use loop to print first 3 elements 
+for idx in range(3):
+    x,y = data_set[idx]
+    print("i = {} Original x = {} Original y = {}".format(idx,x,y))
+    x_a_m,y_a_m = a_m(data_set[idx])
+    print("i = {} Transformed x = {} Transformed y = {}".format(idx,x_a_m,y_a_m))
+
+#create a new data_set object with add_mult object as transform
+cust_data_set = toy_set(transform = a_m)
+
+#use loop to print first 3 elements 
+for idx in range(3):
+    x,y = data_set[idx]
+    print("i = {} Original x = {} Original y = {}".format(idx,x,y))
+    x_a_m,y_a_m = cust_data_set[idx]
+    print("i = {} Transformed x = {} Transformed y = {}".format(idx,x_a_m,y_a_m))

PyTorch_3_Compose_Transforms.py

@@ -0,0 +1,54 @@
+import torch
+from torch.utils.data import Dataset
+from torchvision import transforms
+
+#crate a subclass from Dataset
+class toy_set(Dataset):
+    #contructor
+    def __init__(self,length = 50,transform = None):
+        self.len = length
+        # pylint: disable=E1101 
+        self.x = torch.ones(length,2)
+        self.y = torch.ones(length,1)
+        # pylint: enable=E1101
+        self.transform = transform
+
+    #return data at a given index
+    def __getitem__(self,index):
+        sample = self.x[index],self.y[index]
+        if self.transform:
+            sample = self.transform(sample)
+        return sample
+
+    #return length
+    def __len__(self):
+        return self.len
+
+#crate a transforms class
+class add_mult(object):
+    #contructor
+    def __init__(self,addx = 1,muly = 2):
+        self.addx = addx
+        self.muly = muly
+    #executor
+    def __call__(self,sample):
+        return sample[0] + self.addx , sample[1] * self.muly
+
+class mult(object):
+    def __init__(self,mul = 100):
+        self.mult = mul
+    def __call__(self,sample):
+        return sample[0] * self.mult , sample[1] * self.mult
+
+# Combine the mult() & add_mult()
+data_transform = transforms.Compose([mult(),add_mult()])
+
+#print(data_transform)
+data_set = toy_set()
+composed_data_set = toy_set(transform = data_transform)
+
+for idx in range(3):
+    x,y = data_set[idx]
+    print("i = {} Original x = {} Original y = {}".format(idx,x,y))
+    x_c,y_c = composed_data_set[idx]
+    print("i = {} Transformed x = {} Transformed y = {}".format(idx,x_c,y_c))

PyTorch_4_Image_Datasets.py

@@ -0,0 +1,68 @@
+import torch
+from torch.utils.data import Dataset
+import torchvision.transforms as transforms
+
+import pandas as pd
+import os
+from PIL import Image
+import matplotlib.pyplot as plt
+
+data_dir = ''
+csv_file = 'index.csv'
+""" 
+csv_path = os.path.join(data_dir,csv_file)
+
+data_name = pd.read_csv(csv_path)
+print(data_name.head(5))
+print('File name : {}'.format(data_name.iloc[0,1]))
+print('y : {}'.format(data_name.iloc[0,0]))
+print('Total rows : {}'.format(data_name.shape[0]))
+
+image_name = data_name.iloc[0,1]
+image_path = os.path.join(data_dir,image_name)
+
+image = Image.open(image_path)
+plt.imshow(image,cmap='gray', vmin=0, vmax=255)
+plt.title(data_name.iloc[0, 0])
+plt.show() 
+"""
+class ImageDataset(Dataset):
+    def __init__(self,data_dir,csv_file,transform = None):
+        self.data_dir = data_dir
+        csv_path = os.path.join(self.data_dir,csv_file)
+        self.data_name = pd.read_csv(csv_path)
+        self.len = self.data_name.shape[0]
+        self.transform = transform
+        
+    def __getitem__(self,idx):
+        image_path = os.path.join(self.data_dir, self.data_name.iloc[idx,1])
+        image = Image.open(image_path)
+        y = self.data_name.iloc[idx,0]
+        if self.transform:
+            image =  self.transform(image)
+        return image,y
+
+    def __len__(self):
+        return self.len
+
+""" 
+my_image_dataset = ImageDataset(data_dir = data_dir,csv_file = csv_file)
+
+image,y = my_image_dataset[0]
+plt.imshow(image,cmap='gray', vmin=0, vmax=255)
+plt.title(y)
+plt.show() 
+"""
+
+data_transform = transforms.Compose([transforms.CenterCrop(10), transforms.ToTensor()])
+my_image_dataset = ImageDataset(data_dir = data_dir,csv_file = csv_file,transform=data_transform)
+
+#my_image_dataset = ImageDataset(data_dir = data_dir,csv_file = csv_file,transform=None)
+
+#print(my_image_dataset[0][0].shape)
+
+image,y = my_image_dataset[1]
+print(y)
+plt.imshow(transforms.ToPILImage()(image),cmap='gray', vmin=0, vmax=255)
+plt.title(y)
+plt.show()