Merge pull request #11 from vin00d/main

rel 0.1.2
corazonlabs · Apr 2, 2021 · 5bf3be1 · 5bf3be1
2 parents 54fb129 + 632a5bf
commit 5bf3be1
Show file tree

Hide file tree

Showing 39 changed files with 4,709 additions and 1,555 deletions.
diff --git a/00_basics.ipynb b/00_basics.ipynb
@@ -67,7 +67,8 @@
     "#export\n",
     "from fastai.imports import * \n",
     "from addict import Dict\n",
-    "from datetime import date"
+    "from datetime import date\n",
+    "from collections import OrderedDict"
    ]
   },
   {
@@ -158,19 +159,22 @@
     "        \n",
     "        'SYNTHEA_DATAGEN_DATES' : \n",
     "        {\n",
-    "            '1K'  : date.today().strftime('%d-%m-%Y'),\n",
-    "            '10K' : date.today().strftime('%d-%m-%Y'),\n",
-    "            '20K' : date.today().strftime('%d-%m-%Y'),\n",
-    "            '100K': date.today().strftime('%d-%m-%Y'),\n",
-    "            '250K': date.today().strftime('%d-%m-%Y')\n",
+    "            '1K'  : date.today().strftime('%m-%d-%Y'),\n",
+    "            '10K' : date.today().strftime('%m-%d-%Y'),\n",
+    "            '20K' : date.today().strftime('%m-%d-%Y'),\n",
+    "            '100K': date.today().strftime('%m-%d-%Y'),\n",
+    "            '250K': date.today().strftime('%m-%d-%Y')\n",
     "        },\n",
     "        'CONDITIONS':\n",
     "        {\n",
-    "            'diabetes': '44054006||START',\n",
-    "            'stroke': '230690007||START',\n",
-    "            'alzheimers': '26929004||START',\n",
-    "            'coronary_heart': '53741008||START',\n",
-    "            'lung_cancer': '254637007||START'\n",
+    "            'diabetes': '44054006',\n",
+    "            'stroke': '230690007',\n",
+    "            'alzheimers': '26929004',\n",
+    "            'coronary_heart': '53741008',\n",
+    "            'lung_cancer': '254637007',\n",
+    "            'breast_cancer': '254837009',\n",
+    "            'rheumatoid_arthritis': '69896004',\n",
+    "            'epilepsy': '84757009'\n",
     "        },\n",
     "        'LOG_NUMERICALIZE_EXCEP': True\n",
     "    }\n",
@@ -196,7 +200,7 @@
     "        settings = Dict(settings_template())\n",
     "        Path.mkdir(Path(settings_dir), exist_ok=True)\n",
     "        with open(settings_file, 'w') as s:\n",
-    "            yaml.dump(settings.to_dict(), s, sort_keys=False)\n",
+    "            yaml.dump(settings.to_dict(), s, sort_keys=False, allow_unicode=True)\n",
     "    else:\n",
     "        with open(settings_file, 'r') as s:\n",
     "            settings = Dict(yaml.full_load(s))\n",
@@ -255,7 +259,7 @@
    "id": "buried-scholarship",
    "metadata": {},
    "source": [
-    "- These global variables are used for convenience in many places in the library. They can be over-ridden by passing in different values."
+    "These are global variables with default used for convenience in many places in the library. They can be over-ridden by passing in non-default values where needed."
    ]
   },
   {
@@ -267,11 +271,14 @@
     {
      "data": {
       "text/plain": [
-       "{'diabetes': '44054006||START',\n",
-       " 'stroke': '230690007||START',\n",
-       " 'alzheimers': '26929004||START',\n",
-       " 'coronary_heart': '53741008||START',\n",
-       " 'lung_cancer': '254637007||START'}"
+       "{'diabetes': '44054006',\n",
+       " 'stroke': '230690007',\n",
+       " 'alzheimers': '26929004',\n",
+       " 'coronary_heart': '53741008',\n",
+       " 'lung_cancer': '254637007',\n",
+       " 'breast_cancer': '254837009',\n",
+       " 'rheumatoid_arthritis': '69896004',\n",
+       " 'epilepsy': '84757009'}"
       ]
      },
      "execution_count": null,
@@ -288,9 +295,13 @@
    "id": "floating-uniform",
    "metadata": {},
    "source": [
-    "- These `CONDITIONS` are eventually transformed into **labels** that we will use to train the the deep learning models.\n",
-    "    - The `CONDITIONS` dict is used in pre-processing to identify & label patients with these conditions. \n",
-    "    - The final labels the models train on must be a sub-set of these. "
+    "- These conditions defined in the `CONDITIONS` dictionary are used during pre-processing to identify & label patients that have these conditions\n",
+    "- After pre-processing, a subset of these (some or all of them) are used as labels to train the deep learning models\n",
+    "\n",
+    "\n",
+    "- Thus to train on a different set of labels / conditions\n",
+    "    - First pre-process the dataset using the new conditions\n",
+    "    - And then proceed to training the models"
    ]
   },
   {
@@ -355,11 +366,11 @@
     {
      "data": {
       "text/plain": [
-       "{'1K': '30-03-2021',\n",
-       " '10K': '30-03-2021',\n",
-       " '20K': '30-03-2021',\n",
-       " '100K': '30-03-2021',\n",
-       " '250K': '30-03-2021'}"
+       "{'1K': '04-01-2021',\n",
+       " '10K': '04-01-2021',\n",
+       " '20K': '04-01-2021',\n",
+       " '100K': '04-01-2021',\n",
+       " '250K': '04-01-2021'}"
       ]
      },
      "execution_count": null,
@@ -418,7 +429,7 @@
    "id": "median-christian",
    "metadata": {},
    "source": [
-    "**Please change these paths to defaults in your specific configuration**\n",
+    "**Please change these paths to defaults in your specific configuration** if desired\n",
     "\n",
     "- All of these artifacts need to be in some form of failsafe storage, but not all need to be in version control.\n",
     "- Also, some of them are likely to get big and version control might not be the ideal location (e.g. data, logs and models).\n",