Please share bvlc_reference_caffenet training loss curve

[Jumabek Alikhanov]

Hi,

I am training variant of bvlc_reference_caffenet model,

However, I am getting only 50% accuracy (instead of 57%)

So I want to know if I should continue more iteration. 

For that I need original bvlc_reference_caffenet training loss curve.

Would anybody share their training curve, if they trained?

That would greatly help, since I don't have enough resources to train original model myself.

Below is my training loss curve,

Thanks