Imageai: When will ImageAI be compatible with tensorflow 2.1?

Created on 19 Dec 2019  路  3Comments  路  Source: OlafenwaMoses/ImageAI

Hello,

I am using imageAI with tensorflow 1.14 right now, but since I must have CUDA 10.1 installed, I cannot use my gpu. (because tensorflow 1.14 is not compatible with CUDA 10.1) After a lot of effort I managed to install tensorflow 2.1, which is compatible with CUDA 10.1, but unfortunately, imageAI is not working with tensorflow 2.1 yet. I've read in the documentation that it will be available soon. I was wondering when.

Thanks in advance.

picture bjente

Most helpful comment

@bjente Even if you install 1.4 there are still a bunch of compatibility issues.
I'm using:
python 3.6
tensor-flow 1.4.0
kerras 2.3.1
opencv-python 4.1.2.30

which are all within range according to docs, but I still can't even import ObjectDetection. I think the docs need updating

picture manoadamro on 23 Dec 2019
👍4

All 3 comments

@bjente Even if you install 1.4 there are still a bunch of compatibility issues.
I'm using:
python 3.6
tensor-flow 1.4.0
kerras 2.3.1
opencv-python 4.1.2.30

which are all within range according to docs, but I still can't even import ObjectDetection. I think the docs need updating

manoadamro picture manoadamro on 23 Dec 2019
👍4

I don't think there is a date for this :/ and eah, I agree with you, requirements are not managed/documented very well :/

rola93 picture rola93 on 10 Feb 2020

Same issue here. Would be great if this could work with tensorflow 2.1. difficult to get all the different software packages aligned on versions. I am completely new to python and machine learning and working on this for fun, so take that into account. But thanks to the developers on this project for all the hard work. Not complaining, just providing feedback. Thanks

bellas20 picture bellas20 on 12 Apr 2020
Was this page helpful?
0 / 5 - 0 ratings

Related issues

letchfordnicholas picture letchfordnicholas  路  6Comments
Danielv123 picture Danielv123  路  4Comments
huangl6 picture huangl6  路  6Comments
flozi00 picture flozi00  路  4Comments
Object Detection with 10 lines of code : I did everything as mentioned, but i get the following error. Value Error: Ensure you specified correct input image, input type, output type and/or output image path the input files are there in the same directory.. what am i missing here? Please find the code below : from imageai.Detection import ObjectDetection import os execution_path = os.getcwd() detector = ObjectDetection() detector.setModelTypeAsRetinaNet() detector.setModelPath( os.path.join(execution_path , "resnet50_coco_best_v2.0.1.h5")) detector.loadModel() detections = detector.detectObjectsFromImage(input_image=os.path.join(execution_path , "image.jpg"), output_image_path=os.path.join(execution_path , "imagenew.jpg")) for eachObject in detections: print(eachObject["name"] , " : " , eachObject["percentage_probability"] )
ShobhaShiva picture ShobhaShiva  路  3Comments