Omer Schleifer

thanks for the quick response, the error is in this line: https://github.com/liangfu/mxnet-mobilenet-v2/blob/master/from_mxnet.py#75 compiling the graph. Regarding your suggestion, On my host machine (ubuntu 16.04 , intel i7 cpu) the script...

I'm having the same issue. any suggestions? thanks

@Davidnet - thanks a lot, that's a big help :) one small question, did you build tensorflow with tensroRT?

Thanks, sounds perfect.

Having the same issue, getting ByeBuffer from S3 and passing them as is to http-kit response, cause a memory leak in direct memory. Is there a way to use heap...

@redapple , I am interested in this issue, from what I've seen so far, Scrapy handles graceful shut-down by calling "stop" on CrawlerProcess ,and than waiting for the crawlers themselves...

@kmike - So, If we support a configuration in which the downloader requests buffer is on disk (instead of in memory) that would ensure persistency , right?

I started to implement this as suggested, but I'm unsure about the settings: in the httpcompression class. I can access the global settings by caliing `new Settings()`, but how can...