AlirezaShamsoshoara / Fire-Detection-UAV-Aerial-Image-Classification-Segmentation-UnmannedAerialVehicle

Aerial Imagery dataset for fire detection: classification and segmentation (Unmanned Aerial Vehicle (UAV))
https://ieee-dataport.org/open-access/flame-dataset-aerial-imagery-pile-burn-detection-using-drones-uavs
GNU General Public License v2.0
161 stars 48 forks source link

I did not successfully compile your code #4

Open wangk98 opened 3 years ago

wangk98 commented 3 years ago

Hello, When I run the code of "mode = 'training“, the following error appears. QQ图片20210306171917

Do you know how to solve it?

AlirezaShamsoshoara commented 3 years ago

Hello @wangk98, thanks for raising this issue. This case for the plot only works if you train your model with the METRIC parameters: https://github.com/AlirezaShamsoshoara/Fire-Detection-UAV-Aerial-Image-Classification-Segmentation-UnmannedAerialVehicle/blob/473e289f0f9de127161bedd837567df923e68cd4/training.py#L42-L52

However, in the current version, only, accuracy was used as a metric. Hence, if you make these changes, there won't be any issue: In plotdata.py, change line#21 https://github.com/AlirezaShamsoshoara/Fire-Detection-UAV-Aerial-Image-Classification-Segmentation-UnmannedAerialVehicle/blob/473e289f0f9de127161bedd837567df923e68cd4/plotdata.py#L21

to this:

(fig, ax) = plt.subplots(1, 1, figsize=(13, 13))

and comment out lines# 34 - 45 (inclusive).

I will make these changes in a new branch and I will submit them. Let me know if there is still any issues.

Good luck!

wangk98 commented 3 years ago

Sorry to disturb you again. After I changed it according to the above method, the following error was reported: IMG_20210307_210558 Then I change line#21(in plotdata.py) to (fig, ax) = plt.subplots(2, 1, figsize=(13, 13),squezee=False) the following error was reported: IMG_20210307_205937 and I do not know how to solve it

hyojun-rtdata commented 3 years ago

Hello, thank you for you Great research. I compile your code I have same error.

I changed code and commet lines.

image

cryptedwcoder commented 2 years ago

@hyojun-rtdata @wangk98 @AlirezaShamsoshoara Did you solve it ?

MArslan95 commented 2 months ago

@AlirezaShamsoshoara

Screenshot 2024-05-14 164712

Hi I hope you doing well I ran your code In my system while running. I get the val_loss as nan I tried to find the sol but didn't find anything regarding this so please if you know the reason let me know thank you