Lukket

Fine tune the pretrained alexnet and use it on lfw dataset as a verification protocol

Job Description:

I need to compare the performance of the verification network using pretrained alexnet, however, I found that the Score after fine-tuning is lower than the one without finetuning. I want to find where is wrong.

Færdigheder: Neural Networks, Deep Learning, Pytorch

Om klienten:
( 1 bedømmelse ) Santa Clara, United States

Projekt ID: #36303985

6 freelancere byder i gennemsnit $28 timen for dette job

Zied130

Hello, I will work on this project until I get good results and complete all the tasks. I am prepared to handle projects requiring Deep Learning, Neural Networks skills. Please come over chat and discuss your requir Flere

$30 USD in 7 dage
(8 bedømmelser)
3.6
amirrmusavifr

Hi, I hope you are doing fine. I am full time freelancer and you can get proper and full service from me and I am ready to start work any time. I have almost 4 years of experience in machine learning algorithms. I can Flere

$20 USD på 1 dag
(4 bedømmelser)
3.1
uhasnain

Hi I saw all your code in Deep Learning DL_project3. What I found that you have used multiple dropouts here. which might be a case of your accuracy decrease. This might be one case.Fine-tune parameters are not only. Dr Flere

$30 USD in 7 dage
(0 bedømmelser)
0.0
rezatz1999

Hi, I see your attached notebook and I think freezing some layer and then finetune the model will work better for you. Can you tell me more about your dataset like how many samples do you have for training and testing? Flere

$25 USD in 2 dage
(0 bedømmelser)
0.0
hedibouchelliga

Being a Data scientist with 3 years of experience in the data industry working on different subjects and problems I can deliver the work in 1 business day regards.

$30 USD på 1 dag
(0 bedømmelser)
0.0
CampNeptune

I have worked with AlexNet before. If you trying to some kind of transfer learning task, I can help you with that since I have published research works in this domain.

$30 USD in 2 dage
(0 bedømmelser)
0.0