PROFNET(1)

💖 PROFNET 💖

okay bestie, gather round, because we're about to talk about profnet_* — that's the manual page PROFNET(1) for all you citation girlies — and it is literally neural network implementations in Fortran. yes. Fortran. she's old, she's iconic, she's running your protein predictions, don't @ her. 🧚✨

---

✨ NAME ✨

profnet_* — neural network implementations in Fortran 💅

---

🌸 SYNOPSIS 🌸

profnet_* [OPTION|filePar]

---

🦋 DESCRIPTION 🦋

so the profnet_* binaries are neural network implementations in Fortran, periodt. but here's the ✨lore✨: because of how the code was originally designed, a specific binary gets compiled for each particular network architecture — like, they literally go in and change certain constants in the source code. so there's not one binary to rule them all, there's a binary for every network architecture used. one binary per vibe, essentially. 👑

not me telling you that certain array structures are intentionally indexed out of bounds in some of the binaries 😭 that's not a bug bestie that's a design choice, we don't ask questions, we just accept her the way she is 💀

note bestie, this is important:

This binary should only be used to run with pre-made training data. do not try to use it to train your network — it will produce undesired results, and that's not the slay you think it is. it was made to be used only as part of wrapping (dependent) packages and not as a standalone neural network program. she's a supporting character, not the main character. 🌈

---

🎀 OPTIONS 🎀

heads up: this list is not exhaustive. there's more where this came from, but here's the menu okay?? 🍓

• filePar — the file with input parameters (she also gives you fileIn and fileOut, multitasking queen 💅)

then the numbered ones, in order, because order matters and we'll get to that:

1. "switch"
2. number of input units
3. number of hidden units
4. number of output units
5. number of samples
6. bitacc (typically 100)
7. file with input vectors
8. file with junctions
9. file with output of NN ("none" -> no file written)
10. optional=dbg

• [inter] — will bring up dialog 🫶

---

💜 NOTES 💜

• the 1st MUST be "switch"! like, no negotiation, no exceptions, she's first or the whole thing falls apart fr fr 😤
• tested only with 2 layers! so if you're out here with 3+ layers, you're in uncharted territory bestie, tread carefully 🧚

---

👑 AUTHOR 👑

Burkhard Rost <rost@rostlab.org> 💖

Bug fixes and enhancements by Laszlo Kajan <lkajan@rostlab.org> and Guy Yachdav <gyachdav@rostlab.org> ✨

---

🌈 COPYRIGHT AND LICENSE 🌈

Copyright 1998-2011 by Burkhard Rost <rost@rostlab.org> EMBL, CUBIC (Columbia University, NY, USA) and LION Biosciences (Heidelberg, DE)

Copyright 2009-2011 by Laszlo Kajan <lkajan@rostlab.org> Technical University Munich (Munich, DE)

Copyright 2009-2011 by Guy Yachdav <gyachdav@rostlab.org> CUBIC (Columbia University, NY, USA) and Technical University Munich (Munich, DE)

---

and that's the whole page bestie, no cap 🫶 this was version 1.0.22-, dated 2022-01-19 — literally a whole era, and she's still serving. mwah 💋✨💅