![](\"img/irfu_logo.jpg\"){kind=logo}
\n",
- "\n",
- "----------\n",
- "\n",
- "\n",
- "\n",
- "\n",
- "![\"CMS](\"img/CMS_computing_needs.png\"/)
| Fast/easy coding fast | Fast running | |
|---|---|---|
| Python | ⇔ | C/C++ |
![\"A](\"img/snake-1546033.jpg\")
\n",
- "⇒ Lines of code executed billions of times even for a Kleenex code, written specially for a publication.
" - ] - }, - { - "cell_type": "markdown", - "id": "c7f88121", - "metadata": {}, - "source": [ - "## Python dislikes loop\n", - "\n", - " * A master rule for high-performance code in Python is to avoid writing loop in Python \n", - " + ⇒ push the loop to underlying compiled libraries. Approach of the `numpy` vectorisation.\n", - " \n", - "Let's run a simple loop\n", - " \n", - "[Simple loop in Python](Python-code.ipynb#simple-loop)" - ] - }, - { - "cell_type": "markdown", - "id": "7e87ebd7-31da-4d23-a15f-ea90dc553b6f", - "metadata": {}, - "source": [ - "| Python |\n", - "|--------|\n", - "| 90 ms |" - ] - }, - { - "cell_type": "markdown", - "id": "de449e8a-809f-49ed-92cb-4ef01554d1bf", - "metadata": {}, - "source": [ - "## In C++. Code in `simple-loop.cc`:\n", - "\n", - "```cpp\n", - "#include![\"Plug adaptors\" by dogwelder is licensed under CC BY-NC 2.0](\"img/plugs.jpg\")
![\"Calling](\"img/julia_from_python.jpg\"/)
\n",
- " "
- ]
- },
- {
- "cell_type": "markdown",
- "id": "b272d46e-3395-454b-bfa6-21f0aef9487d",
- "metadata": {},
- "source": [
- "# Let's use Julia for HEP example\n",
- "\n",
- "## CMS dimuon analysis \n",
- "\n",
- " > ⚡ It's an extremely simple analysis, way far from usual LHC analysis\n",
- "\n",
- " > Let's go,\n",
- " \n",
- " * [Dimuon spectrum in Julia](dimu/diMuon-bin-jl.ipynb)\n",
- " "
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "08cb8517",
- "metadata": {},
- "outputs": [],
- "source": []
- },
- {
- "cell_type": "markdown",
- "id": "13e38ba4-d873-4919-821f-fb4502890444",
- "metadata": {},
- "source": [
- "# Conclusions\n",
- "\n",
- "* Julia offers the code running performance of C++, with the same programming language the ease of programming of Python, and a large Ecosystem\n",
- "* Julia is the ideal language for our usage\n",
- "* It's now mature enough to be used for HEP\n",
- " + It's the right time to catch the Julia's train\n",
- " + Needs developers to provides tools, especially for ROOT I/O and a ROOT binding\n",
- " + Needs to be advertised to users. Most of the Physicist does not know it.\n",
- "\n",
- "\n",
- "_Thanks to Jim Pivarski for the tricks to turn out a notebook in a presentations and the nice tutorial he gave on the subject_"
- ]
- }
- ],
- "metadata": {
- "kernelspec": {
- "display_name": "Julia 1.6.1",
- "language": "julia",
- "name": "julia-1.6"
- },
- "language_info": {
- "file_extension": ".jl",
- "mimetype": "application/julia",
- "name": "julia",
- "version": "1.6.1"
- }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}
diff --git a/.ipynb_checkpoints/Julia-dream-PyHEP2021-checkpoint.ipynb b/.ipynb_checkpoints/Julia-dream-PyHEP2021-checkpoint.ipynb
deleted file mode 100644
index 2e8db91..0000000
--- a/.ipynb_checkpoints/Julia-dream-PyHEP2021-checkpoint.ipynb
+++ /dev/null
@@ -1,723 +0,0 @@
-{
- "cells": [
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "8acf72d6-fc3e-439c-9a3f-4db4c942cbac",
- "metadata": {
- "jupyter": {
- "source_hidden": true
- },
- "scrolled": true,
- "tags": []
- },
- "outputs": [],
- "source": [
- "singleCellDisplay = true\n",
- "editor_font_size = 25\n",
- "code_output_font_size = 25\n",
- "markdown_output_font_size = 30\n",
- "markdown_list_font_size = 25\n",
- "HTML(\"\"\"\n",
- "\n",
- "\n",
- "![](\"irfu_logo.jpg\"){kind=logo}
![\"CMS](\"CMS_computing_needs.png\"/)
| Fast/easy coding fast | Fast running | |
|---|---|---|
| Python | ⇔ | C/C++ |
![\"A](\"snake-1546033.jpg\")
\n",
- "⇒ Lines of code executed billions of times even for a Kleenex code, written specially for a publication.
" - ] - }, - { - "cell_type": "markdown", - "id": "c7f88121", - "metadata": {}, - "source": [ - "# Python dislikes loop\n", - "\n", - " * A master rule for high-performance code in Python is to avoid writing loop in Python \n", - " + ⇒ push the loop to underlying compiled libraries. Approach of the `numpy` vectorisation.\n", - " \n", - "Let's run a simple loop\n", - " \n", - "[Simple loop in Python](Python-code.ipynb#simple-loop)" - ] - }, - { - "cell_type": "markdown", - "id": "7e87ebd7-31da-4d23-a15f-ea90dc553b6f", - "metadata": {}, - "source": [ - "| Python |\n", - "|--------|\n", - "| 90 ms |" - ] - }, - { - "cell_type": "markdown", - "id": "de449e8a-809f-49ed-92cb-4ef01554d1bf", - "metadata": {}, - "source": [ - "In C++. Code in `simple-loop.cc`:\n", - "\n", - "```cpp\n", - "#include![\"scared plugs\" by dogwelder is licensed under CC BY-NC 2.0](\"plugs.jpg\")
![\"Calling](\"julia_from_python.jpg\"/)
\n",
- " "
- ]
- },
- {
- "cell_type": "markdown",
- "id": "27bc2a1b-7fe5-4942-884c-1b1741f7a73f",
- "metadata": {},
- "source": [
- "# Calling Python from Julia\n",
- "\n",
- "As simple as calling Julia code"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "461e51f3-647b-4566-891c-6840c5d577da",
- "metadata": {},
- "outputs": [],
- "source": [
- "# Enable Python call:\n",
- "using PyCall\n",
- "\n",
- "# Inport a python module:\n",
- "math = pyimport(\"math\")\n",
- "\n",
- "# Use it as a Julia module:\n",
- "math.sin(math.pi / 4)"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "ce6da619",
- "metadata": {},
- "source": [
- "# Calling C from Julia"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "03fbd939",
- "metadata": {},
- "outputs": [],
- "source": [
- "path = ccall(:getenv, Cstring, (Cstring,), \"SHELL\")\n",
- "unsafe_string(path)"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "8eaf62ef",
- "metadata": {},
- "source": [
- "For C, you will typically write a wrapper in ''Julia'' to handle errors, like:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "497fec73",
- "metadata": {},
- "outputs": [],
- "source": [
- "function getenv(var::AbstractString)\n",
- " val = ccall(:getenv, Cstring, (Cstring,), var)\n",
- " if val == C_NULL\n",
- " error(\"getenv: undefined variable: \", var)\n",
- " end\n",
- " return unsafe_string(val)\n",
- "end"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "409d2d27",
- "metadata": {},
- "outputs": [],
- "source": [
- "println(getenv(\"USER\"))\n",
- "println(getenv(\"SMOKE\")) # ⇒ will through an exception unless you have SMOKE in your environment "
- ]
- },
- {
- "cell_type": "markdown",
- "id": "b272d46e-3395-454b-bfa6-21f0aef9487d",
- "metadata": {},
- "source": [
- "# Let's use Julia for HEP example\n",
- "\n",
- "## CMS dimuon analysis \n",
- "\n",
- " > ⚡ It's an extremely simple analysis, way far from usual LHC analysis\n",
- "\n",
- " > Let's go,\n",
- " \n",
- " * [Dimuon spectrum in Julia](dimu/diMuon-bin-jl.ipynb)\n",
- " * [Dimuon spectrum in Python](dimu/diMuon-bin-py.ipynb)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "08cb8517",
- "metadata": {},
- "outputs": [],
- "source": []
- },
- {
- "cell_type": "markdown",
- "id": "13e38ba4-d873-4919-821f-fb4502890444",
- "metadata": {},
- "source": [
- "# Conclusions\n",
- "....\n",
- "\n",
- "_Thanks to Jim Pivarski for the tricks to turn out a notebook in a presentations and the nice tutorial he gave on the subject_"
- ]
- }
- ],
- "metadata": {
- "kernelspec": {
- "display_name": "Julia 1.6.1",
- "language": "julia",
- "name": "julia-1.6"
- },
- "language_info": {
- "file_extension": ".jl",
- "mimetype": "application/julia",
- "name": "julia",
- "version": "1.6.1"
- }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}
diff --git a/.ipynb_checkpoints/Julia_from_IPython-checkpoint.ipynb b/.ipynb_checkpoints/Julia_from_IPython-checkpoint.ipynb
deleted file mode 100644
index 4a5b03c..0000000
--- a/.ipynb_checkpoints/Julia_from_IPython-checkpoint.ipynb
+++ /dev/null
@@ -1,117 +0,0 @@
-{
- "cells": [
- {
- "cell_type": "markdown",
- "id": "8b4b2ab7",
- "metadata": {},
- "source": [
- "Load the Julia magic extension"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 2,
- "id": "4d4cbc68",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "The julia.magic extension is already loaded. To reload it, use:\n",
- " %reload_ext julia.magic\n"
- ]
- }
- ],
- "source": [
- "%load_ext julia.magic"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "f578aa1c",
- "metadata": {},
- "source": [
- "Excute some code written in ''Julia''"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 3,
- "id": "6fdc193d",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "x = 10 ⇒ x² = 100\n"
- ]
- }
- ],
- "source": [
- "%%julia\n",
- "x = 10\n",
- "y = x^2\n",
- "println(\"x = $(x) ⇒ x² = $(y)\")"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "66b2da7a",
- "metadata": {},
- "source": [
- "Variables defined in Julia can be accessed from Python"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 4,
- "id": "ae1c4b46",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "x = 10 ⇒ x² = 100\n"
- ]
- }
- ],
- "source": [
- "x = %julia x\n",
- "y = %julia y\n",
- "print(f'x = {x} ⇒ x² = {y}')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "241206ff-31f4-4ca2-a93a-783831585db7",
- "metadata": {},
- "outputs": [],
- "source": []
- }
- ],
- "metadata": {
- "kernelspec": {
- "display_name": "Python 3",
- "language": "python",
- "name": "python3"
- },
- "language_info": {
- "codemirror_mode": {
- "name": "ipython",
- "version": 3
- },
- "file_extension": ".py",
- "mimetype": "text/x-python",
- "name": "python",
- "nbconvert_exporter": "python",
- "pygments_lexer": "ipython3",
- "version": "3.9.2"
- }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}
diff --git a/.ipynb_checkpoints/Python-code-checkpoint.ipynb b/.ipynb_checkpoints/Python-code-checkpoint.ipynb
deleted file mode 100644
index ceaf032..0000000
--- a/.ipynb_checkpoints/Python-code-checkpoint.ipynb
+++ /dev/null
@@ -1,72 +0,0 @@
-{
- "cells": [
- {
- "cell_type": "markdown",
- "id": "c3908552-6dee-4767-830a-274c61bf7ee1",
- "metadata": {},
- "source": [
- "\n",
- " \n",
- "# Simple loop example"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 41,
- "id": "17ea0d9f-5134-418d-a057-65bfc5bba7a4",
- "metadata": {},
- "outputs": [],
- "source": [
- "def f():\n",
- " a = 0.\n",
- " for i in range(1, 1000*1000 +1):\n",
- " a = a + 1.0/i\n",
- " print(i)\n",
- " return a"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 42,
- "id": "83dc3089-0828-44a1-a257-27e14f1c01e6",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "1000000\n",
- "14.392726722864989\n",
- "CPU times: user 91.5 ms, sys: 511 µs, total: 92.1 ms\n",
- "Wall time: 90.3 ms\n"
- ]
- }
- ],
- "source": [
- "%%time\n",
- "print(f())\n"
- ]
- }
- ],
- "metadata": {
- "kernelspec": {
- "display_name": "Python 3",
- "language": "python",
- "name": "python3"
- },
- "language_info": {
- "codemirror_mode": {
- "name": "ipython",
- "version": 3
- },
- "file_extension": ".py",
- "mimetype": "text/x-python",
- "name": "python",
- "nbconvert_exporter": "python",
- "pygments_lexer": "ipython3",
- "version": "3.9.2"
- }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}
diff --git a/01-Julia-dream-PyHEP2021.ipynb b/01-Julia-dream-PyHEP2021.ipynb
index 9e8b80d..8ae234d 100644
--- a/01-Julia-dream-PyHEP2021.ipynb
+++ b/01-Julia-dream-PyHEP2021.ipynb
@@ -10,7 +10,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 1,
"id": "a26ab57b-021a-438d-896a-8662e22a44ab",
"metadata": {
"jupyter": {
@@ -30,17 +30,18 @@
" /* resize the markdown output */\n",
" div.jp-MarkdownOutput { font-size: 25px; }\n",
" div.jp-MarkdownOutput li { font-size: 25px; }\n",
- "\n",
- "# /* Shadow around markdown output cells */\n",
- "# div.jp-MarkdownOutput { box-shadow: 0 4px 8px 0 rgba(0, 0, 0, 0.2), 0 6px 20px 0 rgba(0, 0, 0, 0.19);}\n",
- "# div.jp-MarkdownOutput {\n",
- "# padding: 10px 10px 50px 50px;\n",
- "# }\n",
+ " .jp.RenderedHTMLCommon table { font-size: 25px important;}\n",
+ " div.text_cell_render { font-size: 25px; \n",
+ " line-height: 35px;}\n",
+ " div.text_cell_render li { font-size: 25px; }\n",
+ " div.text_cell_render table { font-size: 25px;}\n",
"\n",
" /* Frame around markdown output cells */\n",
- " div.jp-MarkdownOutput { border-style: solid; border-color: #a6a6a6;}\n",
- " div.jp-MarkdownOutput {\n",
- " padding: 10px 10px 50px 50px;\n",
+ " div.jp-MarkdownOutput { border-style: solid; border-color: #a6a6a6;\n",
+ " padding: 10px 10px 50px 50px;\n",
+ " }\n",
+ " div.text_cell_render { border-style: solid; border-color: #a6a6a6;\n",
+ " padding: 10px 10px 50px 50px;\n",
" }\n",
"\n",
"\n",
@@ -50,10 +51,10 @@
"