Karan Purohit comments

Results 37 comments of


                                            Karan Purohit

Classfying the image is taking more time.

Simple.because my repo has darknet plus tesseract. Labels are detected and passed through the tesseract. And darknet is only doing label detection. That's why less time. Didn't you read the...

How to use dvector_create.py

@wrongbattery Have you trained the speaker diarization model using UISRNN completely? If yes pls enlighten us as there is a lot of confusion going on. Can you describe what all...

Multi-GPU training

Yes. It is working on a single GPU. Looking for multi-gpu support. @clarkkev

Multi-GPU training

Is there any plans for multi-gpu support @clarkkev

Training loss

You can always use tensorboard though!

Training loss

I have trained my model on GPU and using tensorboard is similar here. You will find events.out files in your checkpoint folder. Just run tensorboard on it.

Training loss

In tensorboard you get loss and learning rate here I think you can add whatever you want in logs to see them on tensorboard!

Issue while generating pre training data

@stefan-it I changed the vocab to berts format from your provided code snippet. Thanks! One strange thing I found that my training data had a size of around 22 Gb...

Issue while generating pre training data

@parmarsuraj99 Are you training using HF library?

Issue while generating pre training data

Is HF Electra pertaining method available? They were going to publish it.