Batman
Batman
The splits used are the ones provided by the UCF research group. You can find the splits at their website. [UCF101](https://www.crcv.ucf.edu/research/data-sets/ucf101/)
> 1. Browse to c:\program files (x86)\microsoft visual studio\shared\python37_64\lib\site-packages\sc2 (I'm using Python in Visual Studio 2019 environment) > 2. Edit game_data.py and comment out 'assert self.id != 0' with #...
> The file game_data.py ust be there otherwise you will have to copy it from the pysc2 package on github > […](#) > On Sun, Jul 21, 2019 at 3:40...
> 3\. Edit pixel_map.py and comment out 'assert self.bits_per_pixel % 8 == 0, "Unsupported pixel density"' with # in front, save changes. This assert is not in the file
Ok. I got it to work and this is what I did. I am using Python 3.7 and using SC2 4.9.3 1. I force upgraded via this post. [#266 ](https://github.com/Dentosal/python-sc2/issues/266)...
I ran into the same issue. If you reduce the number of classes then the model will converge. For instance, if you reduce ufc101 to 7 classes and then train...
The loss should be computed with the outputs. I have good training now.
> Is there anyone who solve this problem? I also met the same problem. The loss quickly converge ,but the accuracy is only 1% in top 1 and 5% in...
@mithleshmeghwal0 Which code section?
`elliptic.p256Curve` should have no exported fields, right? Because of `p` being lowercase. Inspecting further.