phreakyphoenix / mxnet-gluoncv-aws-coursera

This repo includes my solutions to the Coursera course offered by AWS titled "AWS Computer Vision: Getting Started with GluonCV", in addition to more tutorials and in-depth handson labs. Please :star2: the repo if you like it :point_up: Create an Issue or preferably a PR for any improvement. :rocket:

MXNet-GluonCV-AWS-Coursera

This repo will include some of my solutions to the Coursera course offered by AWS titled "AWS Computer Vision: Getting Started with GluonCV" with some modifications, in addition to more tutorials and in-depth handson labs.

Warning

This is not a solution notebook, so copying the files and submiting won't work and is against Coursera's Honor Code.

a. DO NOT Email me error messages,

b. DO NOT Create an issue in Github, for your own code, such issues will be closed.

c. DO NOT Request edit access to my colab notebooks, to edit/run please use Open in Playgroung mode from the top-left.

Desciption

There are no programming exercises for modules 1 and 2, so it starts from 3 here.

In case the module notebooks aren't loading on your browser due to large file sizes, you can either download them or simply check them out on Jupyter NBviewer (View only) or Colab (View + Edit):

1. Module 3: NBviewer link and Colab link

2. Module 4: NBviewer link and Colab link

3. Module 5: NBviewer link and Colab link

4. Module 6: NBviewer link and Colab link

You don't need to request access for any Colab link, just press Open in Playgroung mode from the top-left to run it.

In all these notebooks, I've not modified the code required to pass the assignmnets, but I've added separate ways to load the data and some personal touches as extra functions at the end of the notebook. Some of the added functionality includes:

1. Testing on multiple images
2. Testing on images uploaded from User.
3. Improved and more accurate visualization based on object detection threshold.

I'd love to merge PRs if you discover any of the content stops working in future or if you have a better idea on how to improve this repo, maybe through tutorials and labs 🔥

mxnet-gluoncv-aws-coursera's Issues

I am facing an error for the module_5 ipynb file

In the show grader output window I am facing an error after the project is being graded. The error is ModuleNotFoundError: No module named 'google' which is showing for this line: "from google.colab import drive". How can I install the google module? I have tried to install by using these commands: "!pip install google" and "!pip install google-colab". But it doesn't work for me.

Assignment AWS computer vision

Question 1
First, let's get a particular dataset for this exercise. We will be using the CIFAR10 dataset which is built-in in MXNet in the gluon.data.vision.datasets library.

In the following cell you will complete the function definition so that it returns the CIFAR10 dataset from MXNet. The function should return both the training and the validation dataset. To make things more exciting, the function should examine both the training and validation dataset and return the index that corresponds to the first occurrence of each class in CIFAR10.

For example, the first class (label) in the CIFAR10 training dataset is 6, this corresponds to index 0 in the training dataset. For the remaining classes in the dataset, find the index in the training dataset where that class first occurs. Repeat the same process for the validation dataset. To summarize, the get_cifar_10_dataset function below should return 4 values. The CIFAR10 training dataset, the indices of the training dataset that correspond to the first occurrence of each class, The CIFAR10 validation dataset and the indices of the validation dataset that correspond to the first occurrence of each class.

def get_cifar10_dataset():
train_data = None
val_data = None

train_indices = {}
val_indices = {}

// code here

I want Solution of practice assignment of week 5.

Question 1
First, let's get a particular dataset for this exercise. We will be using the CIFAR10 dataset which is built-in in MXNet in the gluon.data.vision.datasets library.

In the following cell you will complete the function definition so that it returns the CIFAR10 dataset from MXNet. The function should return both the training and the validation dataset. To make things more exciting, the function should examine both the training and validation dataset and return the index that corresponds to the first occurrence of each class in CIFAR10.

For example, the first class (label) in the CIFAR10 training dataset is 6, this corresponds to index 0 in the training dataset. For the remaining classes in the dataset, find the index in the training dataset where that class first occurs. Repeat the same process for the validation dataset. To summarize, the get_cifar_10_dataset function below should return 4 values. The CIFAR10 training dataset, the indices of the training dataset that correspond to the first occurrence of each class, The CIFAR10 validation dataset and the indices of the validation dataset that correspond to the first occurrence of each class.

CAUTION!: Although the notebook interface has internet connectivity, the autograders are not permitted to access the internet. We have already downloaded the correct models and data for you to use so you don't need access to the internet. Set the root parameter to M5_IMAGES when using a preset dataset. Usually, in the real world, you have internet access, so setting the root parameter isn't required (and it's set to ~/.mxnet by default).

def get_cifar10_dataset():
"""
Should create the cifar 10 network and identify the dataset index of the first time each new class appears

:return: tuple of training and validation dataset as well as label indices
:rtype: (gluon.data.Dataset, dict[int:int], gluon.data.Dataset, dict[int:int])
"""
train_data = None
val_data = None
train_indices = {}
val_indicies = {}

# Use `root=M5_IMAGES` for your dataset

# YOUR CODE HERE

raise NotImplementedError()

return train_data, train_indices, val_data, val_indices

The Assignment given for week 5 is somewhat wrong

aws computer vision module 5

module 5 graded assessment program is throwing an error.so i am unable to submit it.please help me out