def numerical_check(_net, cfg):
net.initialize()
datas = etl(cfg)
bp_loss_f = BP_LOSS_F
loss_function = {}
loss_function.update(bp_loss_f)
from longling.ML.toolkit.monitor import MovingLoss
from longling.ML.MxnetHelper.glue import module
loss_monitor = MovingLoss(loss_function)
# train check
trainer = module.Module.get_trainer(_net,
optimizer=cfg.optimizer,
optimizer_params=cfg.optimizer_params,
select=cfg.train_select)
for epoch in range(0, 100):
for _data in tqdm(datas):
with autograd.record():
bp_loss = fit_f(_net, _data, bp_loss_f, loss_function,
loss_monitor)
assert bp_loss is not None
bp_loss.backward()
trainer.step(cfg.batch_size)
print(eval_f(_net, datas))
print("epoch-%d: %s" % (epoch, list(loss_monitor.items())))
def toolbox_init(
self,
evaluation_formatter_parameters=None,
validation_logger_mode="w",
informer_silent=False,
):
from longling.lib.clock import Clock
from longling.lib.utilog import config_logging
from longling.ML.toolkit.formatter import MultiClassEvalFormatter \
as Formatter
from longling.ML.toolkit.monitor import MovingLoss, \
ConsoleProgressMonitor as ProgressMonitor
self.toolbox = {
"monitor": dict(),
"timer": None,
"formatter": dict(),
}
mod = self.mod
cfg = self.mod.cfg
# 4.1 todo 定义损失函数
# bp_loss_f 定义了用来进行 back propagation 的损失函数，
# 有且只能有一个，命名中不能为 *_\d+ 型
assert self.loss_function is not None
loss_monitor = MovingLoss(self.loss_function)
# 4.1 todo 初始化一些训练过程中的交互信息
timer = Clock()
progress_monitor = ProgressMonitor(
loss_index=[name for name in self.loss_function],
end_epoch=cfg.end_epoch - 1,
silent=informer_silent)
validation_logger = config_logging(
filename=os.path.join(cfg.model_dir, "result.log"),
logger="%s-validation" % cfg.model_name,
mode=validation_logger_mode,
log_format="%(message)s",
)
# set evaluation formatter
evaluation_formatter_parameters = {} \
if evaluation_formatter_parameters is None \
else evaluation_formatter_parameters
evaluation_formatter = Formatter(
logger=validation_logger,
dump_file=mod.cfg.validation_result_file,
**evaluation_formatter_parameters)
self.toolbox["monitor"]["loss"] = loss_monitor
self.toolbox["monitor"]["progress"] = progress_monitor
self.toolbox["timer"] = timer
self.toolbox["formatter"]["evaluation"] = evaluation_formatter