Skip to content
This repository was archived by the owner on Jan 11, 2023. It is now read-only.

Conversation

@DavidFair
Copy link
Contributor

This includes averaging out loaded data (for multiple reference images)
and transmission calculation.

The ROI code is WIP, however the transmission code causes us to run OOM,
so it would be better to figure out a way of resolving that first

One approach at the trade-off of space for speed is to calculate and apply the transmission correction projection by projection, rather than calculate it for the ~2000 projections in one

This includes averaging out loaded data (for multiple reference images)
and transmission calculation.

The ROI code is WIP, however the transmission code causes us to run OOM,
so it would be better to figure out a way of resolving that first
@DavidFair DavidFair added the enhancement New feature or request label Jun 1, 2020
"dark_field = radiography.fits_to_variable(dark_field_dir, average_data=True)\n",
"flat_field = radiography.fits_to_variable(flat_field_dir, average_data=True)\n",
"\n",
"# This blows through 32GB of memory as raw_data[\"sample\"] is already ~17 GB\n",
Copy link
Contributor

@OwenArnold OwenArnold Jun 3, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you implement below in place to avoid memory overhead of temporaries?

sample -= dark
flat -= dark
sample /= flat

or similar

"\n",
"Before starting you must:\n",
"- Ensure that `scipp` and `mantid` are on your `PYTHONPATH`.\n",
"- Generate the `config.py` file using `make_config.py`. Refer to the `README.md` or `python make_config.py --help` for information.\n",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Download data (specify own-cloud location) as a step?

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

enhancement New feature or request

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants