Train the model using new data created in issue #4
Chintan2108 opened this issue · comments
-
Train the model using the new dataset as created in issue #4
-
Compare the model performance, assess the generated images' quality using the stipulated measures (code available in the notebook) and document the same in readme of the PR
-
Update repository readme with the quantitative comparison info of these two models
I would like to work on this
I would also like to contribute to this issue.
@Namyalg @Preetesh21
You both can train the model on the new skewed dataset as sampled by @Preetesh21 in issue #4
- Train the model using the same hyperparameters
- Report analytical and comparative model performance
- Compare your respective models with the current project model (link in readme)
- Compare each others' model
- Update readme with a comparison table in terms of our 3 evaluation metrics - PSNR, SSIM and Pearson Correlation
- Also provide epoch vs error graphs (for both of you)
Okay, will do that
Namya LG
…On Tue, Oct 27, 2020, 6:21 PM vesper2108 ***@***.***> wrote:
@Namyalg <https://github.com/Namyalg> @Preetesh21
<https://github.com/Preetesh21>
You both can train the model on the new skewed dataset as sampled by
@Preetesh21 <https://github.com/Preetesh21> in issue #4
<#4>
- Train the model using the same hyperparameters
- Report analytical and comparative model performance
- Compare your respective models with the current project model
(link in readme)
- Compare each others' model
- Update readme with a comparison table in terms of our 3 evaluation
metrics - PSNR, SSIM and Pearson Correlation
- Also provide epoch vs error graphs (for both of you)
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#5 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AM3BEYPUADSOTYLHXO5QD2LSM265HANCNFSM4SPD74YA>
.