arunm8489

price_data = resources.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index()
# join two dataframes in python: 
data = pd.merge(data, price_data, on='id', how='left')

arunm8489

data['school_state'] = data['school_state'].str.lower()

arunm8489

print(f'Number of null values:{data["teacher_prefix"].isnull().sum()}')

arunm8489

sns.countplot(data['project_is_approved'])
plt.show()

arunm8489

data["teacher_prefix"] = data["teacher_prefix"].fillna('Mrs.')
data["teacher_prefix"] = data["teacher_prefix"].str.replace('.','')
data["teacher_prefix"] = data["teacher_prefix"].str.lower()

arunm8489

def clean_pro(col):
  col = col.str.replace('The', '')
  col = col.str.replace(' ','')
  col = col.str.replace('&','_')
  col = col.str.replace(',','_')
  col = col.str.lower()
  return col


data['project_subject_subcategories'] = clean_pro(data['project_subject_subcategories'])

arunm8489

data['project_grade_category'] = data['project_grade_category'].str.replace(' ','_')
data['project_grade_category'] = data['project_grade_category'].str.replace('-','_')
data['project_grade_category'] = data['project_grade_category'].str.lower()
data['project_grade_category'].value_counts()

arunm8489

import warnings
warnings.filterwarnings("ignore")

import shutil, time
import torch
from torch.utils.tensorboard import SummaryWriter
import torch.nn.functional as F
import torch.nn as nn
from torch.utils.data import TensorDataset, DataLoader
from collections import Counter,defaultdict

arunm8489

data = pd.read_csv('train.csv')
resources = pd.read_csv('resources.csv')
print(f'Train file shape: {data.shape}')
print(f'Resource file shape: {resources.shape}')

arunm8489

# Utility functions

# function to load the classes
def load_classes(class_file):
    fp = open(class_file, "r")
    names = fp.read().split("\n")[:-1]
    return names

# function converting images from opencv format to torch format
def preprocess_image(img, inp_dim):