R&D: Developing a novel approach(ML/DL) for research work
- or -
Post a project like this£2.9k(approx. $3.6k)
- Posted:
- Proposals: 7
- Remote
- #2383783
- Expired
Website Developer, Graphic Designer, Transcriber, Content writer, CAD Expert, Python Developer, Photo Editor, Web Scrapper, JAVA developer, Android developer, Wix/Shopify Expert,
Regensburg
41354296788910369321881624234405725437782719570
Description
Experience Level: Intermediate
All the milestones are contiguous. We will be proceeding to next milestone only when you meet the requirements of the current milestone. The model should be very novel and should be substantial.
Note: In every milestone, I will be working together with you documenting my report. You will be guiding me on writing the research paper. (Things I will documenting: what has been done in previous papers, examining their architecture, devise a new framework can be used and why does the novel method work, what makes this model novel).
Milestone 1:
• Do the literature review of related works in this field (all applications engines, batteries, and
other industrial equipment) and understand the methodologies/proposed work in those papers.
• Come up with the possible limitations/modifications that can be done to the previous best
works.
• Devise the methodology which you think will have the potential to solve the problem and can get
better score.
Milestone 2:
• Preprocess the data
• Implement the novel model which you have devised in Milestone 1.
• Compute the RMSE and try to get the score closer to the best ones at least 14 or 15 in our case as
the highest is been around 11 and other scoring function which the author has provided
Milestone 3:
• Tweak the model to get the best RMSE score (top two or three - around 11) as well scoring function score
to be below 800
• Produce graphs for predicted vs actual.
• Test the score on other 4 similar datasets.
• Pre-process the data sets according to their operating conditions
• Examine the difference in performance and may be modifications.
I will be accepting the milestones only when the model is fairly novel and has reached the RMSE score and other scoring functions requirements.
Note: In every milestone, I will be working together with you documenting my report. You will be guiding me on writing the research paper. (Things I will documenting: what has been done in previous papers, examining their architecture, devise a new framework can be used and why does the novel method work, what makes this model novel).
Milestone 1:
• Do the literature review of related works in this field (all applications engines, batteries, and
other industrial equipment) and understand the methodologies/proposed work in those papers.
• Come up with the possible limitations/modifications that can be done to the previous best
works.
• Devise the methodology which you think will have the potential to solve the problem and can get
better score.
Milestone 2:
• Preprocess the data
• Implement the novel model which you have devised in Milestone 1.
• Compute the RMSE and try to get the score closer to the best ones at least 14 or 15 in our case as
the highest is been around 11 and other scoring function which the author has provided
Milestone 3:
• Tweak the model to get the best RMSE score (top two or three - around 11) as well scoring function score
to be below 800
• Produce graphs for predicted vs actual.
• Test the score on other 4 similar datasets.
• Pre-process the data sets according to their operating conditions
• Examine the difference in performance and may be modifications.
I will be accepting the milestones only when the model is fairly novel and has reached the RMSE score and other scoring functions requirements.
Lalitha B.
0% (0)Projects Completed
-
Freelancers worked with
-
Projects awarded
0%
Last project
20 Apr 2024
United States
New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-
There are no clarification messages.
We collect cookies to enable the proper functioning and security of our website, and to enhance your experience. By clicking on 'Accept All Cookies', you consent to the use of these cookies. You can change your 'Cookies Settings' at any time. For more information, please read ourCookie Policy
Cookie Settings
Accept All Cookies