search

lauzingaretti / DeepGP

A tool to implement Genomic Prediction Experiments using Deep Learning
16 stars 10 forks source link

Installation issue #5

Open vunguyen1907 opened 3 years ago

vunguyen1907 commented 3 years ago

Hi Laura.

I am new to Linux and Python and my below issues could be easy to solve but I try some ways and it did not work. I have a issue during using "pip3 install -r requirements.txt" (pic1). I run ubuntu on Window 10. Then I used Linux system but it still has an issue (pic2).

Anny suggestion on how to solve it? Thanks for your considering.

Vu

pic1 pic2

lauzingaretti commented 3 years ago

hi Vu, in the first image it looks like you are missing a dependence. In fact, it looks like you have a problem with matplotlib but I'm not sure how you could fix it.

Regarding the second image, you are in the main dir DeepGP, however all the programs (CNN and MLPs) are in the subfolder called GP, then I guess you should go to GP folder before running the code, otherwise you should unzip all the files in GP folder into the main directory (DeepGP). Hope it helps you. Note that acc_pearson, etc, are in the main code of CNN (DeepGP/GP/CNN/cnn.py).

Best wishes,

El vie, 19 mar 2021 a las 0:40, vunguyen1907 @.***>) escribió:

Hi Laura.

I am new to Linux and Python and my below issues could be easy to solve but I try some ways and it did not work. I have a issue during using "pip3 install -r requirements.txt" (pic1). I run ubuntu on Window 10. Then I used Linux system but it still has an issue (pic2).

Anny suggestion on how to solve it? Thanks for your considering.

Vu

[image: pic1] https://user-images.githubusercontent.com/79078602/111711197-f01f7f00-8896-11eb-8da1-931a10977f08.png [image: pic2] https://user-images.githubusercontent.com/79078602/111711199-f150ac00-8896-11eb-8597-1f838cabb3c7.png

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/lauzingaretti/DeepGP/issues/5, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACRX2ZMPEZVLTANDVKF64VTTEKFPDANCNFSM4ZNWKQEQ .

-- Laura Zingaretti PhD Student

Genómica de Plantas y Animales https://www.cragenomica.es/es/programas-de-investigacion/genomica-de-plantas-y-animales Statistical and Population Genomics https://www.cragenomica.es/es/grupos-de-invetigacion/statistical-and-population-genomics CENTRE FOR RESEARCH IN AGRICULTURAL GENOMICS (CRAG) Tel. Oficina +34 935636600 ext 3323 Campus UAB - Edifici CRAG | 08193 Cerdanyola | BARCELONA email: @.***

"I grow old ... I grow old" T S E.

"It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so." Mark Twain

vunguyen1907 commented 3 years ago

Hi Laura.

Thanks for your quick responses.

I followed your instructions and probably make it run, however it seem there is persistent issue with the computer, I think. Attached pic is my screenshot. Do you know how to overcome this?

I am using a server computer with Linux systems, 60 cores, 1TB RAM.

Thanks and have a lovely weekend,

Kind regards,

Vu pic3

lauzingaretti commented 3 years ago

Hi Vu,

I've checked this issue in internet and I found it:

The current version of Tensorflow installed via pip uses the AVX instruction set at compile time.This means that your CPU needs to support the AVX instruction set. This instruction set is supported from the second generation of Intel Core CPUs (codenamed SandyBridge). You can compile a Tensorflow from the source that does not use the AVX instruction set. Or find an already compiled one on the internet.

Hope it could helps you

Have a nice weekend

El vie., 19 mar. 2021 23:27, vunguyen1907 @.***> escribió:

Hi Laura.

Thanks for your quick responses.

I followed your instructions and probably make it run, however it seem there is persistent issue with the computer, I think. Attached pic is my screenshot. Do you know how to overcome this?

I am using a server computer with Linux systems, 60 cores, 1TB RAM.

Thanks and have a lovely weekend,

Kind regards,

Vu [image: pic3] https://user-images.githubusercontent.com/79078602/111848066-eb6dd000-8955-11eb-950d-1833d9c2665d.png

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/lauzingaretti/DeepGP/issues/5#issuecomment-803169430, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACRX2ZIYVYP3UYKHUAHZQOTTEPFV3ANCNFSM4ZNWKQEQ .

vunguyen1907 commented 3 years ago

Hi Laura.

Eventually, I can run your example data now with GPU server. Now, I have 2 questions in mind.

First, I see that you use Pearson for evaluating accuracies which means suitable for the continuous trait not categorial or binary trait accuracy assessment? Is DeepGP can handle binary data? if yes, how can I add another metric such as the Mathew coefficient for binary trait?

Second, regarding the application of DeepGP (i.e. for selection the animal), how can I write out the estimated breeding value after running to do that?

Kind regards,

Vu

lauzingaretti commented 3 years ago

Hi Vi,

you could customize the code to add another metric, you can also can use a metric already provided by keras/tf

Regarding the second cuestion, you should use the selected model to predict new outputs. best,

El vie, 26 mar 2021 a las 0:26, vunguyen1907 @.***>) escribió:

Hi Laura.

Eventually, I can run your example data now with GPU server. Now, I have 2 questions in mind.

First, I see that you use Pearson for evaluating accuracies which means suitable for the continuous trait not categorial or binary trait accuracy assessment? Is DeepGP can handle binary data? if yes, how can I add another metric such as the Mathew coefficient for binary trait?

Second, regarding the application of DeepGP (i.e. for selection the animal), how can I write out the estimated breeding value after running to do that?

Kind regards,

Vu

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/lauzingaretti/DeepGP/issues/5#issuecomment-807718548, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACRX2ZMQIZBODGMY2P6UK53TFPBBLANCNFSM4ZNWKQEQ .

-- Laura Zingaretti PhD Student

Genómica de Plantas y Animales https://www.cragenomica.es/es/programas-de-investigacion/genomica-de-plantas-y-animales Statistical and Population Genomics https://www.cragenomica.es/es/grupos-de-invetigacion/statistical-and-population-genomics CENTRE FOR RESEARCH IN AGRICULTURAL GENOMICS (CRAG) Tel. Oficina +34 935636600 ext 3323 Campus UAB - Edifici CRAG | 08193 Cerdanyola | BARCELONA email: @.***

"I grow old ... I grow old" T S E.

"It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so." Mark Twain