MLWhiz

# Dealing with long docs:
max_length = 384 # The maximum length of a feature (question and context)
doc_stride = 128 # The authorized overlap between two part of the context when splitting it 

def prepare_train_features(examples):
    # Tokenize our examples with truncation and padding, but keep the overflows using a stride. This results
    # in one example possible giving several features when a context is long, each of those features having a
    # context that overlaps a bit the context of the previous feature.
    tokenized_examples = tokenizer(
        examples["question" ],

MLWhiz

datasets = load_dataset("squad")
def visualize(datasets, datatype = 'train', n_questions=10):
    n = len(datasets[datatype])
    random_questions=random.choices(list(range(n)),k=n_questions)
    for i in random_questions:
        print(f"Context:{datasets[datatype][i]['context']}")
        print(f"Question:{datasets[datatype][i]['question']}")
        print(f"Answer:{datasets[datatype][i]['answers']['text']}")
        print(f"Answer Start in Text:{datasets[datatype][i]['answers']['answer_start']}")
        print("-"*100)

MLWhiz

from datasets import load_dataset, load_metric
import random
from transformers import AutoTokenizer
import transformers
from transformers import AutoModelForQuestionAnswering, TrainingArguments, Trainer
import torch
from transformers import default_data_collator
from transformers import AutoTokenizer, AutoModelForQuestionAnswering
import torch
tokenizer = AutoTokenizer.from_pretrained("bert-large-uncased-whole-word-masking-finetuned-squad")

MLWhiz

import math
class Complex:
    def __init__(self, re=0, im=0):
        self.re = re
        self.im = im
    def __add__(self, other):
        # If Int or Float Added, return a Complex number where float/int is added to the real part
        if isinstance(other, int) or isinstance(other, float):
            return Complex(self.re + other,self.im)
        # If Complex Number added return a new complex number having a real and complex part

MLWhiz

import math
class Complex:
    def __init__(self, re=0, im=0):
        self.re = re
        self.im = im

    def __add__(self, other):
        if isinstance(other, int) or isinstance(other, float):
            return Complex(self.re + other,self.im)
        elif  isinstance(other, Complex):

MLWhiz

import math
class Complex:
    def __init__(self, re=0, im=0):
        self.re = re
        self.im = im

    def __add__(self, other):
        if isinstance(other, int) or isinstance(other, float):
            return Complex(self.re + other,self.im)
        elif  isinstance(other, Complex):

MLWhiz

import math

class Shape3d:
    def __init__(self, name):
        self.name = name

    def surfaceArea(self):
        pass
        
    def volume(self):

MLWhiz

import math

class Shape:
    def __init__(self, name):
        self.name = name

    def area(self):
        pass
        
    def getName(self):

MLWhiz

num_epochs = 5
for epoch in range(num_epochs):
    model.train()
    for x_batch,y_batch in train_dataloader:
        if train_on_gpu:
            x_batch,y_batch = x_batch.cuda(), y_batch.cuda()
        optimizer.zero_grad()
        pred = model(x_batch)
        loss = loss_criterion(pred, y_batch)
        loss.backward()

MLWhiz

# Whether to train on a gpu
train_on_gpu = torch.cuda.is_available()
print(f'Train on gpu: {train_on_gpu}')# Number of gpus
if train_on_gpu:
    gpu_count = torch.cuda.device_count()
    print(f'{gpu_count} gpus detected.')
    if gpu_count > 1:
        multi_gpu = True
    else:
        multi_gpu = False