From 37229bffe499162bf6159695b18c542fc630eb19 Mon Sep 17 00:00:00 2001
From: Anup Kumar <anup.rulez@gmail.com>
Date: Thu, 25 Feb 2021 15:51:25 +0100
Subject: [PATCH] Update tool

---
 .../default_tensorflow_notebook.ipynb         | 136 ++++++++++++------
 .../interactivetool_ml_jupyter_notebook.xml   |  42 +-----
 2 files changed, 96 insertions(+), 82 deletions(-)

diff --git a/tools/interactive/default_tensorflow_notebook.ipynb b/tools/interactive/default_tensorflow_notebook.ipynb
index 8ce18a1227f..bd238c6c604 100644
--- a/tools/interactive/default_tensorflow_notebook.ipynb
+++ b/tools/interactive/default_tensorflow_notebook.ipynb
@@ -1,9 +1,25 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "disabled-interest",
+   "metadata": {},
+   "source": [
+    "# Welcome to the Galaxy's GPU enabled Interactive Jupyter Lab for Machine Learning."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "resident-livestock",
+   "metadata": {},
+   "source": [
+    "The jupyter notebook is powered by the latest Jupyter Lab (3.0.7) and Tensorflow (2.4.1) which can be used to prototype and develop machine learning and deep learning solutions executing on Galaxy's NVIDIA GPUs. The docker container used can be found at [website](https://github.com/anuprulez/ml-jupyter-notebook/tree/galaxy-inte-docker-ml-jupyter)."
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 1,
-   "id": "several-election",
+   "execution_count": 26,
+   "id": "amber-tunisia",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -12,10 +28,18 @@
     "import matplotlib.pyplot as plt"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "lovely-adoption",
+   "metadata": {},
+   "source": [
+    "### Resources"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 2,
-   "id": "rural-scoop",
+   "execution_count": 27,
+   "id": "central-roller",
    "metadata": {},
    "outputs": [
     {
@@ -31,27 +55,25 @@
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": 3,
-   "id": "binary-blast",
+   "cell_type": "markdown",
+   "id": "available-metro",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Num CPUs Available:  1\n"
-     ]
-    }
-   ],
    "source": [
-    "print(\"Num CPUs Available: \", len(tf.config.list_physical_devices('CPU')))"
+    "### An example neural network created with Tensorflow"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "million-travel",
+   "metadata": {},
+   "source": [
+    "#### Collect data"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
-   "id": "wanted-exception",
+   "execution_count": 28,
+   "id": "nutritional-ghost",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -62,8 +84,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
-   "id": "qualified-realtor",
+   "execution_count": 29,
+   "id": "transparent-equilibrium",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -73,8 +95,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
-   "id": "considerable-private",
+   "execution_count": 30,
+   "id": "adaptive-hanging",
    "metadata": {},
    "outputs": [
     {
@@ -100,8 +122,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
-   "id": "extraordinary-description",
+   "execution_count": 31,
+   "id": "solved-samuel",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -112,8 +134,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
-   "id": "controlling-paste",
+   "execution_count": 32,
+   "id": "embedded-charleston",
    "metadata": {},
    "outputs": [
     {
@@ -139,10 +161,18 @@
     "plt.show()"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "answering-romantic",
+   "metadata": {},
+   "source": [
+    "#### Define neural network architecture"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 9,
-   "id": "coordinate-jefferson",
+   "execution_count": 33,
+   "id": "enormous-guest",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -155,8 +185,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
-   "id": "august-opinion",
+   "execution_count": 34,
+   "id": "architectural-arbor",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -168,7 +198,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "pending-madison",
+   "id": "nasty-difficulty",
    "metadata": {},
    "outputs": [
     {
@@ -176,11 +206,23 @@
      "output_type": "stream",
      "text": [
       "Epoch 1/10\n",
-      "1875/1875 [==============================] - 13s 6ms/step - loss: 0.6163 - accuracy: 0.7844\n",
+      "1875/1875 [==============================] - 11s 6ms/step - loss: 0.6184 - accuracy: 0.7847\n",
       "Epoch 2/10\n",
-      "1875/1875 [==============================] - 11s 6ms/step - loss: 0.3804 - accuracy: 0.8625\n",
+      "1875/1875 [==============================] - 11s 6ms/step - loss: 0.3832 - accuracy: 0.8641\n",
       "Epoch 3/10\n",
-      "1507/1875 [=======================>......] - ETA: 2s - loss: 0.3322 - accuracy: 0.8772"
+      "1875/1875 [==============================] - 10s 5ms/step - loss: 0.3398 - accuracy: 0.8766\n",
+      "Epoch 4/10\n",
+      "1875/1875 [==============================] - 10s 5ms/step - loss: 0.3143 - accuracy: 0.8850\n",
+      "Epoch 5/10\n",
+      "1875/1875 [==============================] - 11s 6ms/step - loss: 0.2972 - accuracy: 0.8895\n",
+      "Epoch 6/10\n",
+      "1875/1875 [==============================] - 11s 6ms/step - loss: 0.2777 - accuracy: 0.8959\n",
+      "Epoch 7/10\n",
+      "1875/1875 [==============================] - 11s 6ms/step - loss: 0.2676 - accuracy: 0.8991\n",
+      "Epoch 8/10\n",
+      "1875/1875 [==============================] - 11s 6ms/step - loss: 0.2572 - accuracy: 0.9029\n",
+      "Epoch 9/10\n",
+      "1418/1875 [=====================>........] - ETA: 2s - loss: 0.2515 - accuracy: 0.9051"
      ]
     }
    ],
@@ -191,7 +233,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "personal-deployment",
+   "id": "designed-billy",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -203,7 +245,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "three-lighting",
+   "id": "random-backup",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -214,7 +256,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "plain-lease",
+   "id": "painful-catch",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -224,7 +266,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "specified-marshall",
+   "id": "accessory-approval",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -263,7 +305,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "prompt-toddler",
+   "id": "characteristic-vegetation",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -279,7 +321,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "naval-concentrate",
+   "id": "promising-mention",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -295,7 +337,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "amended-ethiopia",
+   "id": "hybrid-lover",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -317,7 +359,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "rental-attribute",
+   "id": "demanding-discipline",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -330,7 +372,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "original-architect",
+   "id": "killing-performance",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -343,7 +385,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "color-qatar",
+   "id": "located-picture",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -355,7 +397,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "pretty-channel",
+   "id": "insured-writing",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -366,7 +408,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "small-emission",
+   "id": "waiting-yahoo",
    "metadata": {},
    "outputs": [],
    "source": [
diff --git a/tools/interactive/interactivetool_ml_jupyter_notebook.xml b/tools/interactive/interactivetool_ml_jupyter_notebook.xml
index 34935b75112..611f3bc5279 100644
--- a/tools/interactive/interactivetool_ml_jupyter_notebook.xml
+++ b/tools/interactive/interactivetool_ml_jupyter_notebook.xml
@@ -1,9 +1,9 @@
-<tool id="interactive_ml_tool_jupyter_notebook" tool_type="interactive" name="Interactive Jupyter Notebook for Machine learning" version="0.1">
+<tool id="interactive_ml_tool_jupyter_notebook" tool_type="interactive" name="GPU enabled Interactive Jupyter Notebook for Machine Learning" version="0.1">
     <requirements>
         <container type="docker">docker.io/anupkumar/docker-ml-jupyterlab:galaxy-integration</container>
     </requirements>
     <entry_points>
-        <entry_point name="Jupyter Interactive Tool" requires_domain="True">
+        <entry_point name="GPU enabled Interactive Jupyter Notebook for Machine Learning" requires_domain="True">
             <port>8888</port>
 	    <url>ipython/lab</url>
         </entry_point>
@@ -33,35 +33,7 @@
             chown \$NB_USER ./default_tensorflow_notebook.ipynb &&
             jupyter trust ./default_tensorflow_notebook.ipynb &&
 	    /usr/local/bin/start-notebook.sh
-	##else:
-	    ##set $cleaned_name = re.sub('[^\w\-\.]', '_', str($input.element_identifier))
-            ##cp '$mode.ipynb' ./${cleaned_name}.ipynb &&
-            ##jupyter trust ./${cleaned_name}.ipynb &&
-            ##if $mode.run_it:
-                ##jupyter nbconvert --to notebook --execute --output ./default_tensorflow_notebook.ipynb --allow-errors  ./*.ipynb &&
-            ##else:
-                ##jupyter lab --allow-root --no-browser --NotebookApp.shutdown_button=True &&
-            ###end if
-            ##cp ./default_tensorflow_notebook.ipynb '$jupyter_notebook'
 	#end if
-	
-	##if $mode.mode_select == 'scratch':
-            ## copy default notebook
-	    ##cp '$__tool_directory__/default_notebook_ml.ipynb' ./ipython_galaxy_notebook_ml.ipynb &&
-            ##jupyter trust ./ipython_galaxy_notebook_ml.ipynb &&
-            ##jupyter lab --allow-root --no-browser --NotebookApp.shutdown_button=True &&
-            ##cp ./ipython_galaxy_notebook_ml.ipynb '$jupyter_notebook'
-        ##else:
-            ##set $cleaned_name = re.sub('[^\w\-\.]', '_', str($input.element_identifier))
-            ##cp '$mode.ipynb' ./${cleaned_name}.ipynb &&
-            ##jupyter trust ./${cleaned_name}.ipynb &&
-            ##if $mode.run_it:
-                ##jupyter nbconvert --to notebook --execute --output ./ipython_galaxy_notebook_ml.ipynb --allow-errors  ./*.ipynb &&
-            ##else:
-                ##jupyter lab --allow-root --no-browser --NotebookApp.shutdown_button=True &&
-            ##end if
-            ##cp ./ipython_galaxy_notebook_ml.ipynb '$jupyter_notebook'
-        ##end if
     ]]>
     </command>
     <inputs>
@@ -93,12 +65,12 @@
         </test>
     </tests>
     <help>
-    The Jupyter Notebook is an open-source web application that allows you to create and share documents that contain live code, equations,
-    visualizations and narrative text. Uses include: data cleaning and transformation, numerical simulation, statistical modeling, data visualization,
-    machine learning, and much more.
+    
+    JupyterLab is a next-generation web-based user interface for Project Jupyter. JupyterLab enables you to work with documents and activities such as Jupyter notebooks, text editors, 
+    terminals, and custom components in a flexible, integrated, and extensible manner.
 
-    Galaxy offers you to use Jupyter Notebooks directly in Galaxy accessing and interacting with Galaxy datasets as you like. A very common use-case is to
-    do the heavy lifting and data reduction steps in Galaxy and the plotting and more `interactive` part on smaller datasets in Jupyter.
+    Galaxy offers you to use Jupyter Lab directly in Galaxy accessing and interacting with Galaxy datasets as you like. A very common use-case is to
+    do the heavy lifting such as performing computation on GPUs and data reduction steps in Galaxy and the plotting and more `interactive` part on smaller datasets in Jupyter Lab.
 
     You can start with a new Jupyter notebook from scratch or load an already existing one, e.g. from your collegue and execute it on your dataset.
     If you have a defined input dataset you can even execute a Jupyter notebook in a workflow, given that the notebook is writing the output back to the history.