Frequently Asked Questions
General FAQs
We have various AI projects ranging from Machine Learning (ML), Natural Language Processing (NLP), to Computer Vision (CV).
When you go to the Projects page you will see the list of all available projects. To enroll simply click on the enroll button. Note that you can only be enrolled in one project at a time therefore, you will be able to enroll in another project after you have completed the project you are currently enrolled in.
Each project has its own specific requirements. After you enroll in a project, you will receive a variety of tasks which you will need to complete. Once you submit a task, your supervisor will mark your task as passed or return it to you for revisions. You must pass all the tasks to complete a project.
The deadlines for each task/project are soft deadlines. These deadlines are what your customer is expecting you to meet but your educational institute will be responsible for fixing hard deadlines.
You can only be enrolled in one project at a time. However, if the deadline of a project has passed, you can enroll in a second one while you finish the first project.
There will be one supervisor assigned to you from your own institute. You can contact the supervisor for any required support.
If you are unable to complete a project, you will not be eligible for certification. Prolonged delays in completing tasks may result in the suspension of the project, and repeated failure to meet deadlines could lead to withdrawal from the Practical Training Platform.
Yes, you will be issued a certificate after successfully completing all the enrolled projects and upon the evaluation and approval of the final synopsis by our team.
For technical issues on the platform, please contact us via our support email. For project or task-related queries, kindly reach out directly to your supervisors.
The format of the final synopsis is outlined in the "Final Synopsis" project. You are eligible to submit the synopsis only after completing all the enrolled projects within the allotted time frame.
eAuto FAQs
Injection card 1 and Injection 2 are different. In injection card 1 you are asked to evaluate the dataset quality and prepare a plan to edit the images and in Injection card 2 data preprocessing and augmentation is to be implemented.
Write your explanations in the text cells of a Jupyter Notebook because then you can a) answer all of my questions and b) share any relevant code with its output to support your conclusions or display your actions taken. Please ensure to answer all the questions with supporting code in a Jupyter Notebook; and please remember to submit it as a zip file with the checkpoints file as well.
We understand that you may be more comfortable working in other formats. However, our company prefers submissions in Jupyter Notebook format as it provides the best way to present code, visualizations, and explanations in an integrated and organized manner. This format also ensures consistency across all submissions, streamlining internal workflows and maintaining alignment with the rest of the company. If youre unfamiliar with Jupyter Notebook or facing issues getting started, their official documentation is an excellent resource.
You can always use the internet for assistance, but it is essential to ensure that any code or materials you use are open-source and do not infringe on private licenses or violate copyright laws. Our ethics board strictly prohibits plagiarism or the unauthorized use of proprietary materials. To maintain transparency and integrity, we request that you provide proper references for all resources and materials you use in your work.
As per the guidelines, project tasks must be carried out individually.
The attachment contains the web-scraped data under Injection Card 1 to help you get started. However, we encourage you to expand the dataset further and conduct a comprehensive quality check.
As per the guidelines, the next injection card will be shown after the allotted timeline for the previous task.
The programming language for this project is Python.
As per the guidelines, project tasks must be coded in Python and uploaded in Jupyter Notebooks.
WebHelpers FAQs
As per the limited development infrastructure requirements, you can use any web scraping tool other than the specified ones.
You can make use of any state-of-the-art NLP models for labeling the scraped dataset.
It's up to you to decide upon the type of text summarization to be used. I suggest you try both types and choose the best one.
As per the legalities and pricing structure of third-party companies, you can decide to use third-party APIs to enhance the project's efficiency. However, we suggest researching and exploring open-source libraries for the same functionality.
As per our views, you can use Amazon Lex service that provides APIs to integrate your chatbot with Amazon's platform and enable text or voice interactions.
If you have already attained the data for the project, great! Next, clean the data to make it usable. Data downloaded from the internet contains unnecessary words that need to be removed. Once cleaned, we will guide you on the next steps.
We encourage you to choose the proper method, but as a company, we must perform this task independently. It's better to code this part using models available online.
Our company expects work in Jupyter Notebook format to ensure consistency and integration. If you are unfamiliar, Jupyter Notebook documentation can help you get started.
Data augmentation involves creating variations of your existing data by applying transformations to increase its size and diversity. Here's a detailed guide to get started.
You can use the internet for assistance, but ensure all code is open-source. Provide references for materials used in your submissions.
As per the guidelines, project tasks must be carried out individually.
As per the guidelines, the next injection card will be shown after the allotted timeline for the previous task.
The programming language for this project is Python.
In Injection Card 1, you scraped data for only one product. For Injection Card 2, you need a dataset containing reviews of multiple products.
To avoid this, we recommend not scraping large volumes of data.
Explore libraries like NLTK, Gensim, Hugging Face Transformers, Sumy, and BERT Summarizer for various summarization techniques.
CryptoGuides FAQs
As per the limited development infrastructure requirements, you can use any web scraping tool other than the specified ones.
You can use Chrome Developer tools in your browser by clicking F12. Navigate to the Network tab, filter for Fetch/XHR, and find client-to-server data transmission requests.
You can use a tool called curlconverter to make the required conversions.
You must consider the trading hours of the exchange. Run the data-fetching scripts at intervals as low as possible, but preferably every minute.
In Injection Card 3, build the update function for Bitcoin. Later, in Injection Card 5, build it for Ethereum and Litecoin.
The company requires deployment on Streamlit. However, please research other free deployment platforms and provide a report on your findings.
Refer to the suggestions in Injection Card 1. Evaluate the availability, reliability, and frequency of updates before selection.
Outline the training process, input features, model architecture, and evaluation results for different datasets.
To trade on a live exchange, run the bot on a server, set up an account with an exchange, and configure the bot. However, this is beyond the project scope.
Feel free to create bots for other coins. However, for this project, submit work for Bitcoin, Ethereum, and Litecoin.
This program does not provide financial advice. The goal is to educate users on how bots work, not to make trade recommendations.
Search for data sources that include USD. Many websites provide financial data in multiple currencies.
Ideally, the data should update daily. If a source updates multiple times a day, that’s even better.
Web scraping offers flexibility but has legal and technical challenges. APIs are efficient and structured but may have limits and costs. Choose based on your use case.
Decide based on relevance. Retrain the model with or without specific features to test performance. Document your decisions.
Yes, automation is encouraged. Decide whether updates should be automatic or triggered manually in the deployed app.
Yes, include additional cryptocurrencies. Update the model and add documentation for new currencies.
Use appropriate evaluation metrics for the scenario (e.g., classification vs. regression). Provide your recommendations.
Compare evaluation metrics across models. Present the comparison in a graph or table as suitable.