Pragyan Subedi comments

Results 4 comments of


                                            Pragyan Subedi

Restoring from checkpoint failed. This is most likely due to a Variable name or other graph key that is missing from the checkpoint.

Writing tf.reset_default_graph() before training and saving could have made the checkpoint restorable @coderzjw @caolusg

ValueError: Dimension 1 in both shapes must be equal, but are 100 and 256. Shapes are [64,100] and [64,256].

Please write all tf.concat as follows: z_concat = tf.concat(axis=1, values=[t_z, reduced_text_embedding]) instead of z_concat = tf.concat(1, t_z, reduced_text_embedding) The parameter placement has changed hence specifying the parameters will help. @remyavijeesh22...

Add predict helpers such as: predict_expectation, predict_percentile (similar to lifelines)

Any updates on this issue for implementation? @sebp @konradsemsch

I have some problem for running

@Narak7 change the following line in run.py: dataset = TextDataset(datadir, 64) to dataset = TextDataset(datadir, 512) You can see the mapping in preprocess/dataset.py. While passing the data through run.py make...