search

MaybeShewill-CV / bisenetv2-tensorflow

Unofficial tensorflow implementation of real-time scene image segmentation model "BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation"
https://maybeshewill-cv.github.io/bisenetv2-tensorflow/
MIT License
224 stars 59 forks source link

issues about don't have enough ROM #25

Closed zjj321 closed 4 years ago

zjj321 commented 4 years ago

I don't have enough ROM , so i have to adjust the data_provider in training scripts into /data_provider/cityscapes_reader.

but i dont know how to modify the code in cityscapes_bisenetv2_single_gpu_trainner.py

can u help me?

thank you so much

MaybeShewill-CV commented 4 years ago

@zjj321 https://github.com/MaybeShewill-CV/bisenetv2-tensorflow/blob/34ba8e7e939efa387188510f9eb74cd27f260a8a/data_provider/cityscapes/cityscapes_tf_io.py#L259-L264 Reduce cache size here:)

zjj321 commented 4 years ago

good job，thank you so much

MaybeShewill-CV commented 4 years ago

@zjj321 Welcome:)