License This project is open source and available under the BSD License. Meaning you can copy and commercialise this. :-)

Bookings Predictor For Bikes

A Data Set has been presented to us with information about bike bookings and various associated factors. We are expected to make an ML predictor of future demand so the company can deploy its resources accordingly.

Table of Contents

• General Info
• Technologies Used
• Conclusions
• Acknowledgements

General Information

Target

After surviving the corona associated business crash, the American bike sharing company BoomBikes has hired us to predict forthcoming demand of the bikes to deploy the precious and now diminished resources where the returns can be highest.

Produced Dataset day.csv is in datasets/ directory

Required Outcome

The company wants to know:

1. Variables significant in predicting the demand for shared bikes
2. How well those variables describe the bike demands

let's give them what they want...

Conclusions

Final File Structure,

1. Bike-Sharing-Abhinav-Tyagi.ipynb

Containing Jupyter-Lab With Final Outputs

2. Bike-Sharing-Abhinav-Tyagi.py

Final Jupyter-Lab Notebook Converted to machine executable Python Script.

3. Datasets

day.csv is in datasets/ directory

Technologies Used

Operating System(s)

• Windows 11 Build 22H2 Pro N, 2022, 22621.521m WFEP 100.22634.1000.0
• Debian Based Linux Canonicle Ubuntu 22.04 LTS
• UEFI Dual Boot Setup to switch between Microsoft NT AND UNiX Work Environmnets
• WSL Ubuntu On Windows Used Sparingly

Command Line Interactions

• Windows Terminal 1.15.2713.0
• Windows PowerShell 7.2.6
• Debian Terminal on Ubuntu LTS.

IDEs

• Jupyter-Lab
• Notepad
• Microsoft Visual Studio Code

Package Managers/Version Control Systems

• git (version 2.37.3)
• pip (pip 22.1.2)
• Anaconda -- conda 22.9.0
• apt-get

Programming Languages Used

• Python (3.9.13); (Bundled With conda)
• C (Based Libraries Imported in the Code, Numpy, pandas, etc)

Libraries Used

• Seaborn (0.11.2)
• re (regex interpreter)
• matplotlib (3.5.2)
• numpy (1.21.5)
• pandas(1.4.4)
• nbconvert (6.4.5)

SpreadSheets and Presentation Softwares

• Microsoft Excel 2016
• Microsoft Powerpoint 2016
• Sparingly used Google Sheets

Browsers/Browser Engines

Chromium Based 106.0.5249.91 Browser Brave 1.44.105

Firefox Gecko 103.0.2 / 9 August 2022 Browser Engine.

Communication Technologies

• Airtel Fiber

• Realtek 1000 Megabits/Sec Ethernet Adaptor

• Broadcom WLAN Adaptor

• Google Mail, IMAP for communication between team and client (UpGrad)

Development Machine Hardware

• AMD RYZEN 1600x Hexacore Processor, x64_86 Architect, 6 core 12 threat, 95 Watts TDP.
• Gigabyte B450-M-DS3H Motherboard.
• MBR (Master Boot Record Partitioning) ext4, for Linux, on spinning hard drive.
• GPT NTFS filesystem for Windows on Solid State Drive.
• Final Runs done on WD SATA-III Solid State For Efficiency.
• 16 GiB, tri-channel memory, 2x4GiB and 1x8GiB Crucial DDR4 throttled to 2666MHz clock per slot.
• Gigabye AMD Radeon RX 570, 4GiB Memory GDDR5 7000MHz Clock, 2x 3700 RPM 90mm Blade FANS, PCI-e x16 3.0 32000MiB/s Bridge, Polaris 20 XL (215-0910052)GPU 1244 MHz Clock. 150W Peak Power.
• Corsair SV450, SV Series, 80 Plus Bronze Certified, 450 Watt PSU.

Cloud Services Used

Git/Github. (All other operations and development done on local machine, no cloud services used.)

Instruction Set Linear Algebra Kernel Library

OpenBLAS 0.3.21 (Basic Linear Algebra Subprograms) (Faster Operations on AMD hardware, the default Intel Math Kernel Library is throttled on non-Intel machines. If you aren't using AMD processor, MKL will be faster on your machine.)

---

Acknowledgements

Contact

Created by [@abhinavt0681] - feel free to contact me!

Abhinav Tyagi 
abhinav@vayubiotech.com 
+91-9755504588          

Thanks Upgrad and Lending Club to Give us the Opportunity To Perform this Analysis.