当我运行NN时,进行任何训练的唯一方法是将X除以1000。网络还需要在70000次下以0.03的训练速率进行训练,如果这些值较大,则NN会变差。我认为这是由于数据处理不当以及可能缺乏偏见所致,但我真的不知道。 Code on Google Colab
答案 0 :(得分:1)
简而言之:您提到的所有问题等等。
此外,我建议您花一些时间来学习Python,然后再进行研究。首先,请避免使用namespace Yogabandy2017.StripeWebhook
{
public static class StripeWebhook
{
static string YbDatabaseConnectionString;
static string SendGridApiKey;
static string StripeSecret;
static string StripeApiKey;
static StripeWebhook()
{
SqlProviderServices.SqlServerTypesAssemblyName = typeof(SqlGeography).Assembly.FullName;
YbDatabaseConnectionString = ConfigurationManager.ConnectionStrings["YogaBandyDatabase"].ConnectionString;
SendGridApiKey = ConfigurationManager.ConnectionStrings["SendGridApiKey"].ConnectionString;
StripeSecret = ConfigurationManager.ConnectionStrings["StripeSecret"].ConnectionString;
StripeApiKey = ConfigurationManager.ConnectionStrings["StripeApiKey"].ConnectionString;
}
[FunctionName("StripeWebhook")]
public static async Task<object> Run([HttpTrigger(WebHookType = "genericJson")]HttpRequestMessage req, TraceWriter log)
{
return req.CreateResponse(HttpStatusCode.OK, new
{
greeting = $"Everythings ok!"
});
}
}
}
,如果您不小心,可能会导致无法预料的行为。