14 Days of JINA AI Challenge - Featurepreneur
Hello everyone! I am participating in a 14 days challenge conducted by Featurepreneur on the topic JINA AI. I am excited and happy to work and explore this. I will be sharing my update here throughout this journey.
Day 1 of #14DaysOfLearning with #Jina
Date: 24th October 2021
Learning Updates:
- Attended Orientation
- Researched about Jina (Documentation)
Installation of Jina
Neural Search basics
- A new approach for retrieving information.
- Instead of telling a machine a set of rules to understand what data is what, neural search does the same thing with a pre-trained neural network.
- This means developers don't have to write every little rule, saving them time and headaches, and the system trains itself to get better as it goes along.
- Searches all kinds of unstructured data like images, audio, video, PDF etc.
Three Rudimentary concepts in Jina
- Document - Basic data type in Jina
- Executor - How the documents are processed by Jina
- Flow - How Jina streamlines and scales the Executors
Time Spent: 2 Hrs
Day 2 of #14DaysOfLearning with #Jina
Date: 25th October 2021
Learning Updates:
- Embedding -- The point on a graph where a piece of data fits.
- Build a model and embed pictures into it.
- Neural Search uses embedding(s) from Deep Learning models for similarity search.
- Deep learning models use hundreds of dimensions to understand data and create embedding(s).
- Started with the data collection - building our own dataset.
- Started on the base code for finding similarities.
Time Spent: 4.5 Hrs
Day 3 of #14DaysOfLearning with #Jina
Date: 26th October, 2021
Learning Updates:
- Data Collection
- Data cleanup (Dataset cleaning)
- Dealt with errors while executing the flow section of the code
- Different libraries supported by JinaAI
Time Spent: 2 Hrs
Day 4 of #14DaysOfLearning with #Jina
Date: 27th October 2021
Learning Updates:
- Building dataset
- Learnt how to create a user-defined executor If the length of the text present in the document is > 0, then it returns the document.
- Learnt how to delete a workspace, after creating an index, as the previous index will be stored in the workspace document.
Time Spent: 1 Hrs
Day 5 of #14DaysOfLearning with #Jina
Date: 28th October 2021
Learning Updates:
- Went through selenium and beautiful soup for scraping data for our datasets
- Faced some errors in flow and executor and tried solving them
- Started the implementation of searching similarity based on the sample datasets
- Streamlit basics
- Link for Streamlit Documentation: streamlit.io
- Fastest way to build and share data applications
- Streamlit is an open-source Python library that makes it easy to create and share beautiful, custom web apps for machine learning and data science.
Time Spent: 1.5 Hrs
Day 6 of #14DaysOfLearning with #Jina
Date: 29th October 2021
Learning Updates:
- Scrapped some data
- Dataset cleaning
- Went through the documentation
- Understood the basics
Source:
Time Spent: 2 Hrs
Day 7 of #14DaysOfLearning with #Jina
Date: 30th October 2021
Learning Updates:
- Advantages of JINA: --> It's semantic --> Data type - agnostic --> Program can receive data in multiple formats or from multiple sources --> Wide array of models
Makes neural search so much easier to work on
- Challenges which can be overcome using JINA: --> Infrastructure --> Optimization
Source:
Time spent : 1.5 Hrs
Day 8 of #14DaysOfLearning with #Jina
Date: 31st November 2021
Learning Updates:
- DocumentArrray --> A DocumentArray is a list of Document objects. You can construct, delete, insert, sort and traverse a DocumentArray like a Python list.
- JINA can search: text, images, audio, video, 3d mesh, proteins
- Went through JINA AI's git repository
Source:
Time Spent: 2 Hrs
Day 9 of #14DaysOfLearning with #Jina
Date: 1st November 2021
Learning Updates:
- Went through some videos to understand the working of jina ai
- Tried solving an error encountered with flow
- Learnt how to use streamlit
Source:
Time Spent: 2 Hrs
Day 10 of #14DaysOfLearning with #Jina
Date: 2nd November 2021
Learning Updates:
- Made some changes to the base code
- Went through JINA AI's repository issues to find similar cases
- Worked on flow error
- Github link: https://github.com/IshitaG-2002IGK/lyrics-jina
Source:
Time Spent: 2 Hrs
Day 11 of #14DaysOfLearning with #Jina
Date: 3rd November 2021
Learning Updates:
- Collected data
- Learned more about flow and it's concepts
- Learnt more about document array
- Learnt about the driver in Jina
- Went through JINA AI's repository issues to find similar cases
Worked on flow error
Github link: (github.com/IshitaG-2002IGK/lyrics-jina)
Time Spent: 2 Hrs
#Featurepreneur #JINAAI #Featurethon #JINA14DaysChallenge #Featurepreneur #JINAAI #neuralnetworks #machinclearning
Day 12 of #14DaysOfLearning with #Jina
Date: 4th November 2021
Learning Updates:
- Made some changes to the base code
- meddled with errors and as usual tried to fight them away
- Learnt about various Document object attributes of jina
- Viewed the jina course and understood the deeper meaning of the concepts to it
- Went through JINA AI's repository issues to find similar cases and examples
- Github link: (github.com/IshitaG-2002IGK/lyrics-jina)
Time Spent: 3 Hrs
#Featurepreneur #JINAAI #Featurethon #JINA14DaysChallenge #Featurepreneur #JINAAI #neuralnetworks #machinclearning
Day 13 of #14DaysOfLearning with #Jina
Date: 5th November 2021
Learning Updates:
- Completed the project
Working on Visualization
Github link: (github.com/IshitaG-2002IGK/lyrics-jina)
Time Spent: 3 Hrs
Day 14 of #14DaysOfLearning with #Jina
Date: 6th November 2021
Learning Updates:
- Making last updates for our project
Visualization graphs
Github link: (github.com/IshitaG-2002IGK/lyrics-jina)
Time Spent: 4 Hrs
#Featurepreneur #JINAAI #Featurethon #JINA14DaysChallenge #NeuralNetwork #DeepLearning