Anshoo Mehra

@mfeblowitz Thanks for reporting and debugging this. I am running into the same issue, I am just curious why only few of us running into this issue and based on...

If this helps anyone: I started with data file size over 3G with over 7 million sentences, VM was running out of RAM after a couple hours of run (...

> Hi Anshoo, > > Thank you so much for your kind words! > > 1. We are currently working on BERT support for activation exploration. We don't have immediate...

@HenryDashwood great work, I have been struggling on BART Summarization to ONNX & this gives me a ray of hope... [ My earlier tries to onnx never gave final inference...

**Resolved** " RuntimeError: THPVariable_Check(tuple_elem) INTERNAL ASSERT FAILED " by upgrading to torch 1.8.1

@HenryDashwood et al am I reading it right: 1. Outcome of your analysis is that ONNX and Quantized are slower than PyTorch ?? 2. If so, what is the best...

@HenryDashwood appreciate the acknowledgment! Yes, please do continue (if time permits), yours is the only implementation that I have seen working without losing the final output quality & if ONNX...

@HenryDashwood I have been able to make some progress with ONNX, however, shifting to GPUs seems challenging. I have made ONNX runtime working to engage GPUs using provider=['CUDAExecutionProvider'] (This step...