Anshuman Mohan
Anshuman Mohan
Makes progress towards #1673: brings our systolic array development into the modern era of eDSL usage.
This PR makes a little progress towards #1673. I have used a handful of existing helpers from the builder library when the answer was really very clear, but have not...
Our [PIFO](https://github.com/calyxir/calyx/blob/main/calyx-py/test/correctness/queues/pifo.py) is implemented rather peculiarly, and here I propose we rethink it. ### Present A PIFO orchestrates two [FIFO](https://github.com/calyxir/calyx/blob/main/calyx-py/test/correctness/queues/fifo.py)s. When an item is `push`ed into the PIFO, it runs...
Right now the eDSL lets you generate pretty free-wheeling Calyx, even if that Calyx is clearly wrong. I wonder how people feel about making the eDSL a little cleverer. Below...
The algorithm calls `parse_data.py`, which in turn calls odgi's Python bindings, to learn the input graph's dimensions. Instead, use `mygfa.preprocess`: https://github.com/cucapra/pollen/blob/de4f8dbee35003d2ce95d2df4b0300f56b4c0a54/mygfa/mygfa/preprocess.py#L64-L69 This will free us from odgi's Python bindings. This...
Back at MemPan23, we heard the following refrain a few times: > Pangenomics is one of those fields that have been "solved" by the discovery of the right data structure....
At the root of Pollen we have [`process.py`](https://github.com/cucapra/pollen/blob/main/process.py) and I'm not sure if it's meant to be there. @susan-garry would you mind taking a quick look?
The PR https://github.com/cucapra/pollen/pull/116 does some work towards getting rid of `odgi`'s Python bindings, but leaves testing for future work. This is an issue to track that. Closing this issue will...
The PR https://github.com/cucapra/pollen/pull/116 does some work towards getting rid of odgi's Python bindings, but leaves documentation for future work. This is an issue to track that. Closing this issue will...
See full discussion here: https://github.com/cucapra/pollen/discussions/79