laptop with 4 GB RAM and Core i5 can train R-CNN ?

[akrsrivastava]

It can train.Might take a few months.

[shaheryarCS]

😄😄😄😄

On Fri, 3 Jul 2020 at 22:14, akrsrivastava notifications@github.com wrote:

It can train.Might take a few months.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/EdjeElectronics/TensorFlow-Object-Detection-API-Tutorial-Train-Multiple-Objects-Windows-10/issues/498#issuecomment-653627313, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKB2ML3WHSCG5FLI7NMWDALRZYGY3ANCNFSM4N5EGRMA .

[ZarifRamzan]

If you dont have a powerful computer, Dont worry you can use Google Colab to train your model.

https://medium.com/analytics-vidhya/training-an-object-detection-model-with-tensorflow-api-using-google-colab-4f9a688d5e8b

https://medium.com/@moshe.livne/training-tensorflow-for-free-pet-object-detection-api-sample-trained-on-google-collab-c2e65f4a9949

[ZarifRamzan]

You can download model zoo coco here: https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md

[bm777]

Hello to every one, Am training a bird specie object detection. I have 6 classes(with 525 image, 75picturesper class) and using:

software and libraire

• tensorflow1.15
• python3.6

model:

• faster_rcnn_resnet152_pets.config

Laptop:

• OMEN 15 which have RTX graphic card (6Go) on Ubuntu 18.04.

The level: loss:1.4247278, step: 75185 The loss still decreasing but slowly.

I started the training since yesterday and I stopped the training and continued at this morning at the saved checkpoint as you can know. But at this stage, I saved and export the model.*-7108 to test test prediction, but i did not get prediction and no boxes was drawn, only if I decrease the treshold from 0.6 to 0.1, i get some false and true prediction.

My question is: does that means my model was not learned very well? or should I continue the training until I reach 200k step.

[bm777]

An now still learning: