Installation issues with pip about chmpy HOT 5 CLOSED

Thanks Peter. I was using pip and python that point to the same conda environment. I had a few follow up questions: 1. The pip install git… command does work and seems to install the code fine with just one hiccup. “ Successfully built chmpy Installing collected packages: pytz, zipp, tzdata, six, pyparsing, pillow, packaging, numpy, MarkupSafe, kiwisolver, fonttools, cycler, trimesh, scipy, python-dateutil, Jinja2, importlib-resources, contourpy, pandas, matplotlib, seaborn, chmpy ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. scikit-learn 1.2.2 requires threadpoolctl>=2.0.0, which is not installed. Successfully installed Jinja2-3.1.2 MarkupSafe-2.1.3 chmpy-1.1 contourpy-1.2.0 cycler-0.12.1 fonttools-4.44.1 importlib-resources-6.1.1 kiwisolver-1.4.5 matplotlib-3.8.1 numpy-1.26.2 packaging-23.2 pandas-2.1.3 pillow-10.1.0 pyparsing-3.1.1 python-dateutil-2.8.2 pytz-2023.3.post1 scipy-1.11.3 seaborn-0.13.0 six-1.16.0 trimesh-4.0.4 tzdata-2023.3 zipp-3.17.0 WARNING: There was an error checking the latest version of pip. “ So, I had to manually pip install threadpoolctl for chmpy to work. 2) Another request: can you update the file here https://github.com/peterspackman/chmpy/blob/main/tests/acetic_acid.png. It seems to not exist. 3) On a slightly different note, do you have examples/templates that shows the basic lines of code that will generate the 2D di-de Hirshfeld surface of a particular pair atomic contacts? Also, wondering how a QM engine is scripted in here to generate the optimized contacts or if it is still fine to just feed in an already QM-optimized CIF files.

Your assumption is correct - I meant fingerprint plots indeed. So, your pointers on generating this plot does answer my question. If you are able to provide any example lines of code on generating this fingerprint for specific atom pairs, it’d be super helpful to characterize differences between different polymorphs of molecular crystals. Thanks for the other comments as well. Thanks, Satish From: Peter Spackman ***@***.***> Date: Tuesday, November 14, 2023 at 9:26 PM To: peterspackman/chmpy ***@***.***> Cc: Satish Kumar Iyemperumal (he/him/his) ***@***.***>, Author ***@***.***> Subject: Re: [peterspackman/chmpy] Installation issues with pip (Issue #4) CAUTION: This email originated from outside of Vertex. Do not open links or attachments unless you know the sender and believe it is safe. If you suspect that this email is a potentially malicious email, please click on the Report Phish button. 1. I'll try and figure out why those dependencies aren't resolved properly. My guess is it's to do with packages that are already installed. 2. I have updated the README to point to the correct file location. 3. I'm not quite sure what you exactly mean here. So I'll break down as I think there's been some confusion: Are you trying to generate fingerprint plots? I think you may have some wires crossed between what a Hirshfeld surface is as they are not the same thing (though they are related). I'm going to assume this is the case as that's very likely. There's no builtin single method to do it, but an example is here: from chmpy import Crystal from chmpy.crystal.fingerprint import fingerprint_histogram, plot_fingerprint_histogram # change to your crystal structure file c = Crystal.load("src/chmpy/tests/test_files/acetic_acid.cif") # the default separation is only 0.2, but for a histogram this is a bit sparse # obviousl you can tweak it depending on your use case surfaces = c.hirshfeld_surfaces(separation=0.1) for i, surf in enumerate(surfaces): hist = fingerprint_histogram(surf) plot_fingerprint_histogram(hist, filename=f"surface_{i}.png") Should look something like the below: [Image removed by sender. surface_0]<https://urldefense.com/v3/__https:/user-images.githubusercontent.com/7966525/282984222-65a8c00f-26ba-4f00-8dbc-b8550963c073.png__;!!FRfS_D4!ZrOj5gzrBrOKaL2qdjmk2kGqTpiN3A4zpqV3nO1qZrgcUdOQP5yExSujX5BMRzpW8M2Ox7K_D24J9y4PG7fd991WRSnqLeH4Aek$> Naturally you can tweak this however you want with number of bins, etc. just take a look at the code in src/chmpy/crystal/fingerpint.png it's pretty straightforward If you're trying to generate an actual 2D Hirshfeld surface (more of an isoline) then there's no built in code to do that, but you'd just have to evaluate the StockholderWeight function along the plane of interest. As for the second part, there is no QM engine scripted in here? There's nothing really QM about the Hirshfeld surface, $d_e$, $d_i$ etc. The surfaces is done using the promolecule approximation i.e. sum of non-interacting spherical atoms and always has been. It's fine to use whatever CIF you wish to make surfaces for. Also keep in mind if you're not doing this for molecules (or molecular crystals) you'll have to dig into the machinery a bit more as the convenience methods are only available for atomic environments or molecular environments. — Reply to this email directly, view it on GitHub<https://urldefense.com/v3/__https:/github.com/peterspackman/chmpy/issues/4*issuecomment-1811702877__;Iw!!FRfS_D4!ZrOj5gzrBrOKaL2qdjmk2kGqTpiN3A4zpqV3nO1qZrgcUdOQP5yExSujX5BMRzpW8M2Ox7K_D24J9y4PG7fd991WRSnqhmovR_o$>, or unsubscribe<https://urldefense.com/v3/__https:/github.com/notifications/unsubscribe-auth/BD6CXHD6OX3BXL7NQXWDP4LYEQR5HAVCNFSM6AAAAAA7JZZ2B6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMJRG4YDEOBXG4__;!!FRfS_D4!ZrOj5gzrBrOKaL2qdjmk2kGqTpiN3A4zpqV3nO1qZrgcUdOQP5yExSujX5BMRzpW8M2Ox7K_D24J9y4PG7fd991WRSnqbjsPFF0$>. You are receiving this because you authored the thread.Message ID: ***@***.***> This email message and any attachments are confidential and intended for use by the addressee(s) only. If you are not the intended recipient, please notify me immediately by replying to this message, and destroy all copies of this message and any attachments. Thank you.