Learn practical skills, build real-world projects, and advance your career

What all can you find in a biology cell?

Human protein classification assignment of zerotogans pytorch freecodecamp course

Biology : study of living organisms is full of mysteries!
There are things that you can see with naked eyes like a caterpillar feed on mulberry leaves, you garden shrub turing into a giant tree and listen your own heartbeat. But there are some things too for which you Superman's vision like stomata of leaves, ganlia of earthworms (eewww) striation of skeletal muscles.
OOF! Too much jargon flying around
But yeah in crux, we wish to see some amazing things which occur at microscopic level and eventually unravel the mystery of human existence
So, to take a baby step in this heraculean mission, we can to identify the cell bodies that occur in human cell, like nucleus - the root of cell replication, mitochrondia - the power of cell and few more
Lets check out them below!

#import some important tool for our mission
import os
import torch
import pandas as pd
import numpy as np
from torch.utils.data import Dataset, random_split, DataLoader
from PIL import Image
import torchvision.models as models
import matplotlib.pyplot as plt
from sklearn.metrics import f1_score
import torchvision.transforms as tt
import torch.nn.functional as F
import torch.nn as nn
from torchvision.utils import make_grid
%matplotlib inline
Main_Dir = '../input/jovian-pytorch-z2g/Human protein atlas'

Train_Dir = Main_Dir + '/train'                           # Contains training images
Test_Dir = Main_Dir + '/test'                             # Contains test images

Train_Csv = Main_Dir + '/train.csv'                       # Contains real labels for training images
Test_Csv = '../input/jovian-pytorch-z2g/submission.csv'   # File to our result
train_df = pd.read_csv(Train_Csv)
train_df.head()

Hmm... After first look at the dataset we see that it has image and in stead of textual labels we see some numbers.
however we know what these number present. For the given dataset we can only identify 10 such cell bodies and those are