DeepLearning

Is the vocabulary of BERT the same as the vocabulary of BERT-joint?

1

As mentioned in the [technical report,](https://arxiv.org/pdf/1901.08634.pdf) special markup tokens, such as "[Paragraph=N]" and "[Table=N]", were introduced. I think there are no such tokens in the vocabulary for BERT model. So...

How many images used for each batch?

1

I was trying to understand the get_minibatch function in /lib/roi_data_layer/minibatch.py. The assert statement in the function always spit an error (assert len(im_scales) == 1, "Single batch only";assert len(roidb) == 1,...

when is the self.buid method of the HyperModel class defined

1

I was trying to understand this [self._build = self.build](https://github.com/keras-team/keras-tuner/blob/d55c7e453cec94860a93e41b75d3e188b0f30af5/kerastuner/engine/hypermodel.py#L45). I could not understand when self.build is defined.

The reason for using "[MASK]" to replace a masked word in the LIME method

1

In the implementation of LIME class, the [mask_string is [MASK]](https://github.com/PAIR-code/lit/blob/2f5eb09cf0d1a61ad4e59caca33c301af03ec64d/lit_nlp/components/lime_explainer.py#L65). This is also a TODO note, `#TODO(lit-dev): make configurable in UI`. So it seems a user can use other token...

An error when running examples /FinRL_Ensemble_StockTrading_ICAIF_2020.ipynb

5

I was running the example in the title, and run into this error "If using all scalar values, you must pass an index". The error traces back to the code...

had a hard time to understand how negative log likelihood is calculated in the regression example

3

I was trying to understand "loss = tf.reduce_mean(0.5*tf.log(output_sig_pos) + 0.5*tf.div(tf.square(y - output_mu),output_sig_pos))". Could you elaborate more on this?

Data to run the code

1

I am so glad to find your work, and finally see a nn that beats tree models. Would it possible for you to upload the "/data2/yanjianhuan/research/data/tabular/data" to cloud? So it...