利用RNN网络来自动生成python代码
-
修改
config.py文件# train path TRAIN_PATH = [ # '/Users/zijiao/tensorflow', # 此处添加待训练的代码路径 ]
-
运行
train.py开始训练,输出如下
训练完模型之后,直接运行coding.py即可开始自动生成代码,如下
Load last model params successfully. Start predict... def test_URLTddresviound(shouldConffts(saventer): cada = Resition('Helidenent') self.assertEqual(n_moded.add(self.username, reactor) return None c.name = self.mktype("somethod') bNd[0], B_MESSD = IBool(1,4) self.container.remactorAuttom_optswith(14, makeR), (True) cmptraties in type() cleed = port return ClientLogChecker().statil_netxrForm(self.assertFalseCicneLogoint, self, *args)