I have an issue when trying to profile my model. It showed that it can't disable profiler when it is not running.
I have tried many ways, but I could not find how to fix that error. Could someone give me a hint on could I proceed with this?
P/s: Below is the trace of my error. I used fastai library for training the neural network.
RuntimeError Traceback (most recent call last)
<ipython-input-11-5fc263f447e9> in <module>
----> 1 lrn.lr_find()
2 lrn.recorder.plot(suggestion=True)
~\Anaconda3\lib\site-packages\fastai\train.py in lr_find(learn, profiling, start_lr, end_lr, num_it, stop_div, wd)
41 cb = LRFinder(learn, start_lr, end_lr, num_it, stop_div)
42 epochs = int(np.ceil(num_it/len(learn.data.train_dl))) * (num_distrib() or 1)
---> 43 learn.fit(epochs, profiling, start_lr, callbacks=[cb], wd=wd)
44
45 def to_fp16(learn:Learner, loss_scale:float=None, max_noskip:int=1000, dynamic:bool=True, clip:float=None,
~\Anaconda3\lib\site-packages\fastai\basic_train.py in fit(self, epochs, profiling, lr, wd, callbacks)
209 else: self.opt.lr,self.opt.wd = lr,wd
210 callbacks = [cb(self) for cb in self.callback_fns + listify(defaults.extra_callback_fns)] + listify(callbacks)
--> 211 fit(epochs, profiling, self, metrics=self.metrics, callbacks=self.callbacks+callbacks)
212
213 def create_opt(self, lr:Floats, wd:Floats=0.)->None:
~\Anaconda3\lib\site-packages\fastai\basic_train.py in fit(epochs, profiling, learn, callbacks, metrics)
108 print(epoch)
109 with torch.autograd.profiler.profile(use_cuda=False) as prof:
--> 110 loss = loss_batch(learn.model, xb, yb, learn.loss_func, learn.opt, cb_handler)
111 else:
112 loss = loss_batch(learn.model, xb, yb, learn.loss_func, learn.opt, cb_handler)
~\Anaconda3\lib\site-packages\torch\autograd\profiler.py in __exit__(self, exc_type, exc_val, exc_tb)
260 if not self.enabled:
261 return
--> 262 records = torch.autograd._disable_profiler()
263 self.function_events = EventList(parse_cpu_trace(records))
264 return False
RuntimeError: can't disable profiler when it's not running```
