Vision Models in Google Colab
$15-25 USD / time
Buget:
$150
Steps:
1)
Find 5 vision models that does semantic segmentation.
Could be from:
[login to view URL]
[login to view URL]
[login to view URL], or other sources.
Important: models should be recent (from the last 9 years: 2015-2024), well known and having an research article about the model with results. Because we are going to benchmark them and get new results. Make sure all found models can be executed and re-trained in the free version of Google Colab using GPU environment. Models should not be too big or complex, it should comply with Google Colab free version limits.
2)
Create a free version of Google Colab notebook using GPU environment. In notebook do the following:
2.1) change each model's architecture found at step 1, by appending(merging) our small tensorflow model:
input_layer = custom_input_layer()
hidden = [login to view URL](10, activation='gelu')(input_layer)
hidden = [login to view URL]()([input_layer, hidden])
hidden = [login to view URL](10, activation='gelu')(hidden)
2.2) original models found at step 1, should preserve its original weights and parameters, because we just extend the original models by merging them with a part of another model. Make sure the secondary model (ours) preserve its original weights as well.
2.3) The secondary model (ours) could be appended at the begging of found model or at the end of found model. Make a function that have a option to allow us to chose how to append the model (begging, end).
2.4) When extending the found model, the resulting model should have our model as a separate branch. The input layer of Our model should receive a tensor of pixels, same input data that found model uses for input. final layer of our model should be connected using layer concatenation with found model's layer.
2.5) make sure that all models can be merged with our tensorflow model, and the modified model will be tested using same evaluation metrics for all models. This is like a benchmark where we check if our model improved the results of existing(modified) models by extending the found models with our model. So we have to add/merge our model to it, and then retrain the resulting model same way as original found model.
Note:
There are 2 models that we have to connect: found model and ours model. When connecting/merging two models, we create a new branch, and must give the same input to our model which forms a branch. Then output of our model/branch is concatenated with a layer from found model. So we just extend the found model by adding our model to it. When you connect our model to found model make sure that you do it correctly without errors.
3)
Train again the models on their original datasets which previously were trained from their original paper. We should compare the results of original model and modified model. In notebook do the following:
3.1) Train each modified model from step 2 on their original dataset
3.2) Collect all their results in similar format as original model's paper
Payment:
Only after project was completed 100% and delivered on time. You agree not to be paid for partial/incomplete project or if deadline was respected.
Projekt ID: #37945614
Om projektet
133 freelancere byder i gennemsnit $28/timen for dette job
Hi, I hope you are doing fine. I have almost 10 years of experience in machine learning algorithms. I can implement various types of artificial intelligence algorithms including yours with Matlab, Python and etc. I hav Flere
I'll find recent, well-known semantic segmentation models, adapt them in a Google Colab notebook with GPU, and extend their architectures. Then, I'll retrain them on original datasets and compare results with the origi Flere
Hi there. ⭐⭐ I am a full stack Developer ⭐⭐ with experience in HTML, Javascript, Bot Developement, Web Developement, MY SQL, Machine Learning (ML), PYTHON, Deep Learning and Communications. I have worked with many sim Flere
Hi , I'm sure that I can do this job. I'm artificial intelligence engineer experienced in Computer Vision, Data Science, Machine/Deep Learning. I have accomplished many projects like yours, Also I will arrange with you Flere
Hello There, I can do this fast. This a piece of cake to me. Lets chat to discuss deadline Agree with your Budget 150$ You can visit my Profile https://www.freelancer.com/u/ExpertSoul
Hello, good time Hope you are doing well I'm expert in MATLAB/Simulink, Python, HTML5, CSS3, Java, JavaScript and C/C#/C++ programming and by strong mathematical and statistical background, have good flexibility for s Flere
Hello, With over a decade's worth of experience as a full-stack web and mobile developer, I'm well-versed in the crucial technologies required to implement your vision effectively. That includes Python, which is essen Flere
As both an eager and experienced web developer with a deep understanding of Machine Learning and Python, I believe I'm uniquely positioned to complete your project. With HTML 5, Css 3, Javascript, Typescript in my tool Flere
Hi there, I'm thrilled to apply for your Vision Models in Google Colab project. With 4-5 years of experience in Pytorch, Machine Learning (ML), Deep Learning, Python and Tensorflow, I'm confident in my ability to brin Flere
Dear Client, I have extensive experience in Python, Machine Learning (ML), Tensorflow, Pytorch, and Deep Learning, as evidenced by my successful portfolio and positive customer reviews. I am confident in my ability to Flere
Hello. I read your requirement i will do that. Please come on chat we will discuss more about this. I will waiting your reply.
Hi. Thanks for your invitation I have just read your proposal and I am sure I can complete the project on time. I am an expert in ML/DL who has many years of experiences. Please contact me to discuss the project in mor Flere
Hi, Sergiu S.! I have worked with similar projects so that I can provide you with a satisfied result. So I will share you my previous work similar with your project via chatting. Having confirmed the job posting "V Flere
Dear , I am excited to submit my proposal for your project, "Vision Models in Google Colab." My expertise in Python, Machine Learning (ML), Tensorflow, Pytorch, and Deep Learning makes me well-equipped to handle the t Flere
Hello. I've read your description. As you can see in my portfolio, I've already developed a generative AI project and also developed segmentation model using yolo. I like to use colab to run that python models and the Flere
