Loaded pretrained model but train accuracy starts from zero,

[KhurramPirov]

Description

I loaded pretrained model, but got zero train accuracy (acc) at the same time pretrained model has 80% accuracy.
Can I do like this?

`sym, arg_params, aux_params = mx.model.load_checkpoint(os.path.join(args.pretrained, "model"), args.pretrained_epoch)
sym = get_symbol(args)
model = mx.mod.Module(symbol=sym, label_names=None, context=ctx)
model = mx.mod.Module(symbol=sym, context=ctx, label_names=None)
model.bind(for_training=True, data_shapes=[('data', (600, 3, 112, 112))],
label_shapes=model._label_shapes)

  model.set_params(arg_params, aux_params, allow_missing=False)`

Error:
RuntimeError: simple_bind error. Arguments:
data: (100, 3, 112, 112)
Error in operator one_hot0: [15:14:19] src/operator/tensor/./indexing_op.h:1193: Check failed: (*in_attrs)[0] != -1 (-1 vs. -1) Index type must be set for one_hot operator

[keerthanvasist]

@mxnet-label-bot add [python, question]

[keerthanvasist]

Hey @KhurramPirov, thank you for asking the question.
For questions, please submit them on the MXNet discussion forum (https://discuss.mxnet.io) where it will get a wider audience and allow others to learn as well.
Thanks!

[braindotai]

Below code should work

model.save_checkpoint('model', epoch = 0)

sym, arg_params, aux_params = mx.model.load_checkpoint('model', epoch = 0)

model = mx.mod.Module(sym, context = mx.gpu())
model.bind(data_shapes = train_data.provide_data, label_shapes = train_data.provide_label)
model.set_params(arg_params, aux_params)

[KhurramPirov]

It doesn't work inside batch_callback and model.fit. пт, 12 июл. 2019 г. в 14:08, Rishik Mourya <notifications@github.com>:

…

Below code should work model.save_checkpoint('model', epoch = 0) sym, arg_params, aux_params = mx.model.load_checkpoint('model', epoch = 0) model = mx.mod.Module(sym, context = mx.gpu()) model.bind(data_shapes = train_data.provide_data, label_shapes = train_data.provide_label) model.set_params(arg_params, aux_params) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <https://github.com/apache/incubator-mxnet/issues/15496?email_source=notifications&email_token=AH2IALWXSFZS2HGDQC5XVTLP7BQ3HA5CNFSM4H7EN2K2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZZOQQQ#issuecomment-510847042>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AH2IALUDS2I4EIDGSB4CI5TP7BQ3HANCNFSM4H7EN2KQ> .

-- Khurram Pirov skype: matrixrextended mob: +7-916-439-53-76

[KhurramPirov]

Hi! I attached an example for mnist_code, just train for some checkpoint and then try to load it. All in the notebook, I can't get the same accuracy after loading from checkpoints. Please check my notebook. The same bug is in my main code. пт, 12 июл. 2019 г. в 14:08, Rishik Mourya <notifications@github.com>:

…

Below code should work model.save_checkpoint('model', epoch = 0) sym, arg_params, aux_params = mx.model.load_checkpoint('model', epoch = 0) model = mx.mod.Module(sym, context = mx.gpu()) model.bind(data_shapes = train_data.provide_data, label_shapes = train_data.provide_label) model.set_params(arg_params, aux_params) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <https://github.com/apache/incubator-mxnet/issues/15496?email_source=notifications&email_token=AH2IALWXSFZS2HGDQC5XVTLP7BQ3HA5CNFSM4H7EN2K2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZZOQQQ#issuecomment-510847042>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AH2IALUDS2I4EIDGSB4CI5TP7BQ3HANCNFSM4H7EN2KQ> .

-- Khurram Pirov skype: matrixrextended mob: +7-916-439-53-76

[KhurramPirov]

And pyfile пн, 15 июл. 2019 г. в 15:21, Пиров Хуррам Хуршедович < khurram.pirov@phystech.edu>:

…

Hi! I attached an example for mnist_code, just train for some checkpoint and then try to load it. All in the notebook, I can't get the same accuracy after loading from checkpoints. Please check my notebook. The same bug is in my main code. пт, 12 июл. 2019 г. в 14:08, Rishik Mourya ***@***.***>: > Below code should work > > model.save_checkpoint('model', epoch = 0) > > sym, arg_params, aux_params = mx.model.load_checkpoint('model', epoch = 0) > > model = mx.mod.Module(sym, context = mx.gpu()) > model.bind(data_shapes = train_data.provide_data, label_shapes = train_data.provide_label) > model.set_params(arg_params, aux_params) > > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub > <https://github.com/apache/incubator-mxnet/issues/15496?email_source=notifications&email_token=AH2IALWXSFZS2HGDQC5XVTLP7BQ3HA5CNFSM4H7EN2K2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZZOQQQ#issuecomment-510847042>, > or mute the thread > <https://github.com/notifications/unsubscribe-auth/AH2IALUDS2I4EIDGSB4CI5TP7BQ3HANCNFSM4H7EN2KQ> > . > -- Khurram Pirov skype: matrixrextended mob: +7-916-439-53-76

-- Khurram Pirov skype: matrixrextended mob: +7-916-439-53-76

[braindotai]

Here is your working solution notebook. Ask me anything if you've further doubts.(Yes I'm mouryarishik)