I am attempting to train a structured data model based off the Rossmann notebook. I have created the learner using
m = md.get_learner(emb_szs, len(df.columns)-len(cat_vars),
0.04, 1, [1000, 500], [0.001, 0.001], y_range=y_range)
m.lr_find()
but m.lr_find() is giving the following error:
~/fastai/courses/dl1/fastai/model.py in fit(model, data, n_epochs, opt, crit, metrics, callbacks, stepper, swa_model, swa_start, swa_eval_freq, **kwargs)
109 if not isinstance(data, Iterable): data = [data]
110 if len(data) == 1: data = data * len(n_epochs)
--> 111 for cb in callbacks: cb.on_phase_begin()
112 model_stepper = stepper(model, opt.opt if hasattr(opt,'opt') else opt, crit, **kwargs)
113 ep_vals = collections.OrderedDict()
AttributeError: 'LR_Finder' object has no attribute 'on_phase_begin'
I have tried to debug the error and found that cb (LR_Finder) has on_batch_begin, on_batch_end, on_epoch_end etc etc but no on_phase_begin function.
My data looks like the following:
emb_szs:
[(38, 19),
(8, 4),
(2, 1),
(3, 2),
(32, 16),
(7, 4),
(25, 13),
(61, 31),
(61, 31),
(3, 2),
(3, 2),
(3, 2),
(3, 2),
(2, 1),
(2, 1)]
len(df.columns)
186
len(cat_vars)
15
y_range
(0, 11.023686.....)