Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add illustration of inhomogeneity from absorption #113

Open
wants to merge 9 commits into
base: main
Choose a base branch
from
Open

Add illustration of inhomogeneity from absorption #113

Changes from 2 commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
53343f5
add illustration of inhomogeneity from absorption
jokasimr Dec 10, 2024
c3c0a60
fix: move visualization to separate notebook
jokasimr Dec 17, 2024
3401b79
fix: text in notebook
jokasimr Jan 20, 2025
aee7dc8
Merge branch 'main' into display-transmission
jokasimr Jan 20, 2025
1822947
docs: add to index
jokasimr Jan 23, 2025
f361d77
Merge branch 'display-transmission' of github.com:scipp/essdiffractio…
jokasimr Jan 23, 2025
1cf402a
docs: change title and add disk section
jokasimr Jan 23, 2025
de85d7d
fix
jokasimr Jan 23, 2025
bd985c9
Merge branch 'main' into display-transmission
jokasimr Jan 23, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
88 changes: 72 additions & 16 deletions docs/user-guide/dream/dream-visualize-absorption.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@
"id": "0",
"metadata": {},
"source": [
"# DREAM instrument view\n",
"# Visualize effect of absorption\n",
"\n",
"This notebook is an example of how to use the `compute_transmission_map` function to visualize the effect of (sample) absorption on the intensity in the detector.\n",
"\n",
Expand Down Expand Up @@ -53,31 +53,41 @@
"source": [
"dg = dream.io.load_geant4_csv(dream.data.get_path(\"data_dream0_new_hkl_Si_pwd.csv.zip\"))[\"instrument\"]\n",
"dg = sc.DataGroup({key: detector[\"events\"] for key, detector in dg.items()})\n",
"dg"
"dg.keys()"
]
},
{
"cell_type": "markdown",
"id": "4",
"metadata": {},
"source": [
"## Display inhomogeneity from absorption\n",
"## Inhomogeneity from absorption in mantle detector\n",
"\n",
"If the sample absorbs or scatters a large fraction of the incoming neutrons, the intensity in the detector will vary even if the scattering is inhomogeneous.\n"
]
},
{
"cell_type": "markdown",
"id": "5",
"metadata": {},
"source": [
"### Rod-like sample oriented along x-axis"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "5",
"id": "6",
"metadata": {},
"outputs": [],
"source": [
"material = Material(\n",
"# The material determines the rate of absorption and scattering\n",
"vanadium = Material(\n",
" scattering_params=ScatteringParams.for_isotope('V'),\n",
" effective_sample_number_density=sc.scalar(0.07192, unit='1/angstrom**3')\n",
")\n",
"sample_shape = Cylinder(\n",
"# The shape determines the shape and the orientation of the sample\n",
"rod_shape = Cylinder(\n",
" radius=sc.scalar(1, unit='cm'),\n",
" height=(height := sc.scalar(5., unit='cm')),\n",
" # Cylinder is oriented along the x-axis\n",
Expand All @@ -89,13 +99,13 @@
{
"cell_type": "code",
"execution_count": null,
"id": "6",
"id": "7",
"metadata": {},
"outputs": [],
"source": [
"transmission_fraction_mantle = compute_transmission_map(\n",
" sample_shape,\n",
" material,\n",
" rod_shape,\n",
" vanadium,\n",
" beam_direction=sc.vector([0, 0, 1]),\n",
" wavelength=sc.linspace('wavelength', 4, 8, 20, unit='angstrom'),\n",
" # To make it faster, don't compute the transmission fraction for every detector pixel, it's not necessary for the visualization.\n",
Expand All @@ -105,29 +115,75 @@
"\n",
"# The visualization expects the `position` coord to denote detector positions, and `wavelength` to be the last dimension.\n",
"transmission_fraction_mantle.coords['position'] = transmission_fraction_mantle.coords.pop('detector_position')\n",
"dream.instrument_view(transmission_fraction_mantle.transpose((*set(transmission_fraction_mantle.dims) - {'wavelength'}, 'wavelength')), dim='wavelength', camera=Camera(position=sc.vector((-1.5, 1.5, -1.5), unit='m'), look_at=sc.vector((1, -1/3, 1/2), unit='m')))"
"dream.instrument_view(transmission_fraction_mantle.transpose((*set(transmission_fraction_mantle.dims) - {'wavelength'}, 'wavelength')), dim='wavelength', camera=Camera(position=sc.vector((-2, 2, -2), unit='m'), look_at=sc.vector((1, -1/3, 1/2), unit='m')))"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "7",
"id": "8",
"metadata": {},
"outputs": [],
"source": [
"transmission_fraction_mantle"
]
},
{
"cell_type": "markdown",
"id": "9",
"metadata": {},
"source": [
"## Disk-like sample oriented along x-axis"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "8",
"id": "10",
"metadata": {},
"outputs": [],
"source": [
"disk_shape = Cylinder(\n",
" radius=sc.scalar(5, unit='cm'),\n",
" height=(height := sc.scalar(1., unit='cm')),\n",
" # Cylinder is oriented along the x-axis\n",
" symmetry_line=(symmetry_line := sc.vector([1, 0, 0])),\n",
" center_of_base=-height * symmetry_line / 2,\n",
")\n",
"\n",
"transmission_fraction_mantle = compute_transmission_map(\n",
" disk_shape,\n",
" vanadium,\n",
" beam_direction=sc.vector([0, 0, 1]),\n",
" wavelength=sc.linspace('wavelength', 4, 8, 20, unit='angstrom'),\n",
" # To make it faster, don't compute the transmission fraction for every detector pixel, it's not necessary for the visualization.\n",
" detector_position=dg['mantle'].coords['position']['strip', ::4]['wire', ::2].copy(),\n",
" quadrature_kind='cheap',\n",
")\n",
"\n",
"# The visualization expects the `position` coord to denote detector positions, and `wavelength` to be the last dimension.\n",
"transmission_fraction_mantle.coords['position'] = transmission_fraction_mantle.coords.pop('detector_position')\n",
"dream.instrument_view(transmission_fraction_mantle.transpose((*set(transmission_fraction_mantle.dims) - {'wavelength'}, 'wavelength')), dim='wavelength', camera=Camera(position=sc.vector((-2, 2, -2), unit='m'), look_at=sc.vector((1, -1/3, 1/2), unit='m')))"
]
},
{
"cell_type": "markdown",
"id": "11",
"metadata": {},
"source": [
"## Does absorption influence the intensity in the endcap detector?"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "12",
"metadata": {},
"outputs": [],
"source": [
"transmission_fraction_endcap = compute_transmission_map(\n",
" sample_shape,\n",
" material,\n",
" rod_shape,\n",
" vanadium,\n",
" beam_direction=sc.vector([0, 0, 1]),\n",
" wavelength=sc.linspace('wavelength', 4, 8, 20, unit='angstrom'),\n",
" detector_position=dg['endcap_backward'].coords['position']['strip', 0].copy(),\n",
Expand All @@ -140,10 +196,10 @@
},
{
"cell_type": "markdown",
"id": "9",
"id": "13",
"metadata": {},
"source": [
"As expected the impact of absorption on the inhomgenity of the intensity is not nearly as large for the endcap detector as it is for the mantle detector."
"As expected the impact of absorption on on the intensity in the endcap detector is close to uniform."
Copy link
Member

Choose a reason for hiding this comment

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

It doesn't look so uniform when you look at the figure... until you look at the colorbar.
Screenshot_20250123_142736

Should we use some common colorbar limits for all 3 views?

Also, do we really have to use the heavy instrument view, or could we use 2d scatter plots instead?
You could make z-y plots for the mantle, and x-y plot for the end cap?

The 3d views dramatically increase the size of the docs, and thus the repository itself.
It would also mean that you don't need to mess around with Camera settings...

Copy link
Contributor Author

@jokasimr jokasimr Jan 23, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

Should we use some common colorbar limits for all 3 views?

No I don't think that's important. It's better to keep it simple and to make it look like something they might encounter in their own notebook.

The 3d views dramatically increase the size of the docs, and thus the repository itself.

I didn't know that, why does it increase the size of the repo?
I see what you mean.

Did you check how much the repo size is increased?

We could make it 2d, it's not good to make the repo cumbersome to download. Byt the 3d view is nicer to play around with.

Copy link
Contributor Author

@jokasimr jokasimr Jan 24, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

Took a look at the sizes now, after building the docs locally I have the following files:

-rw-rw-r-- 1 johannes johannes  17M jan 23 09:26 dream-instrument-view.html
-rw-rw-r-- 1 johannes johannes  18M jan 23 09:25 dream-instrument-view.ipynb
-rw-rw-r-- 1 johannes johannes 4,5M jan 23 09:26 dream-visualize-absorption.html
-rw-rw-r-- 1 johannes johannes 4,8M jan 23 09:25 dream-visualize-absorption.ipynb

So it seems the absorption visualization notebook is significantly smaller than the instrument view notebook.

I also tried cloning essdiffraction to check how fast it was to download:

~/crap $ git clone https://github.com/scipp/essdiffraction.git
Cloning into 'essdiffraction'...
Receiving objects: 100% (3761/3761), 9.63 MiB | 10.66 MiB/s, done.

Compressed it seems the size (of the current repo) is closer to 10Mb.

Overall I get the impression that it doesn't really matter if we put this file here, it's unlikely to make the repo hard to download for anyone.

Copy link
Member

Choose a reason for hiding this comment

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

10Mb is a lot. Maybe we should really reduce the size of the instrument view notebook?

the 3d view is nicer to play around with.

It might be nice to play with, but I don't think it adds any information to what we are viewing? I think the views might be clearer in 2D. I will have a go at making them and we can compare.

Copy link
Contributor Author

@jokasimr jokasimr Jan 27, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

  1. 10Mb is not a lot, it takes 2s to download.
  2. This change is easily revertible. If we notice slowness we can update the docs and redeploy.

It might be nice to play with, but I don't think it adds any information to what we are viewing?

Let's just ask Celine if this solves her use-case? Then we don't have to guess and re-implement something that already works.

]
}
],
Expand Down