Merge pull request #6 from torchmd/fixes

supplement for the description
torchmd · Jun 21, 2023 · edeecf1 · edeecf1
2 parents b9e6c61 + e35ae3f
commit edeecf1
Show file tree

Hide file tree

Showing 3 changed files with 13 additions and 2 deletions.
diff --git a/README.md b/README.md
@@ -1,6 +1,6 @@
 # Machine Learning Coarse-Grained Potentials of Protein Thermodynamics
 
-This repository contains code, data and tutarial for reproducing the paper "Machine Learning Coarse-Grained Potentials of Protein Thermodynamics". 
+This repository contains code, data and tutarial for reproducing the paper "Machine Learning Coarse-Grained Potentials of Protein Thermodynamics". https://arxiv.org/abs/2212.07492
 
 ## Background
 

diff --git a/Tutorials/1-Tutorial-CG-NNP-Training.ipynb b/Tutorials/1-Tutorial-CG-NNP-Training.ipynb
@@ -1,6 +1,7 @@
 {
  "cells": [
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -15,6 +16,7 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -33,6 +35,7 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -52,6 +55,7 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -102,13 +106,15 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "# Training"
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -203,7 +209,7 @@
     "```\n",
     "where `$PATH` is the path to your `torchmd-net` repo.\n",
     "\n",
-    "The training saves 8 best epochs. The progress of the training is saved in TensorBoard session. The training and validation curves for the training of full Chignolin dataset are presented here:"
+    "The training saves 8 best epochs. The progress of the training is saved in TensorBoard session. The training requires a GPU and should take a couple of hours on a standard GPU."
    ]
   }
  ],

diff --git a/Tutorials/2-Tutorial-CG-Simulation.ipynb b/Tutorials/2-Tutorial-CG-Simulation.ipynb
@@ -1,6 +1,7 @@
 {
  "cells": [
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -24,13 +25,15 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "# Simulation "
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -83,6 +86,8 @@
     "\n",
     "where `$PATH` is a path to your `torchmd` repo (not `torchmd-net` linke in `1-Tutorial-CG-NNP-Training.ipynb`)\n",
     "\n",
+    "The simulation GPU and should take a couple of hours on a standard GPU, however can be also run on a CPU.\n",
+    "\n",
     "Now we will go through options in a configuration file:\n",
     "\n",
     "* First we define `device`, the directory `log_dir` and name of the output files `output`.\n",