Skip to content

feat: sensibility and specificty metrics #10

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
rcsm89 wants to merge 2 commits into
base: main
Choose a base branch
from
Open

feat: sensibility and specificty metrics #10

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
450093a
feat: sensibility and specificty metrics
rcsm89 Feb 2, 2024
b832d5a
refactor: using snow instead of sklean
rcsm89 Feb 5, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
154 changes: 94 additions & 60 deletions notebooks/2_snowml.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,10 +1,10 @@
{
"nbformat": 4,
"nbformat_minor": 2,
"metadata": {},
"cells": [
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import warnings\n",
"\n",
Expand All @@ -15,87 +15,87 @@
"from snowflake.ml.utils.connection_params import SnowflakeLoginOptions\n",
"from snowflake.snowpark import Session\n",
"from snowflake.snowpark import types as T\n",
"from snowflake.snowpark.functions import col\n",
"from snowflake.snowpark.functions import col, sum as snow_sum\n",
"\n",
"warnings.simplefilter(action=\"ignore\", category=UserWarning)"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"session = Session.builder.configs(SnowflakeLoginOptions()).getOrCreate()"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"titanic_df = session.table(\"titanic\")"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"titanic_df.show()"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Columns with null values and their respective counts\n",
"null_counts = [\n",
" (col_name, titanic_df.where(col(col_name).isNull()).count())\n",
" for col_name in titanic_df.columns\n",
"]\n",
"null_counts"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"titanic_df = titanic_df.drop([\"AGE\", \"DECK\", \"ALIVE\", \"ADULT_MALE\", \"EMBARKED\"])"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"titanic_df = titanic_df.withColumn(\"FARE\", titanic_df[\"FARE\"].astype(T.FloatType()))\n",
"\n",
"titanic_df.show()"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"cat_cols = [\"SEX\", \"CLASS\", \"WHO\", \"EMBARK_TOWN\"]\n",
"num_cols = [\"PCLASS\", \"SIBSP\", \"PARCH\", \"FARE\"]"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"impute_cat = SimpleImputer(\n",
" input_cols=cat_cols,\n",
Expand All @@ -106,13 +106,13 @@
"\n",
"titanic_df = impute_cat.fit(titanic_df).transform(titanic_df)\n",
"titanic_df.show()"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"OHE = OneHotEncoder(\n",
" input_cols=cat_cols,\n",
Expand All @@ -124,22 +124,22 @@
"\n",
"titanic_df = OHE.fit(titanic_df).transform(titanic_df)\n",
"titanic_df.show()"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"train_df, test_df = titanic_df.random_split(weights=[0.8, 0.2], seed=8)"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"xgb = XGBClassifier(\n",
" input_cols=train_df.drop(\"SURVIVED\").columns,\n",
Expand All @@ -149,22 +149,45 @@
"\n",
"# Train\n",
"xgb.fit(train_df)"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"result = xgb.predict(test_df)"
],
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"execution_count": null
"source": [
"def specificity_score(df, y_true_col_names, y_pred_col_names):\n",
" tn = snow_sum((col(y_true_col_names) == 0) & (col(y_pred_col_names) == 0)).alias(\"TN\")\n",
" fp = snow_sum((col(y_true_col_names) == 0) & (col(y_pred_col_names) == 1)).alias(\"FP\")\n",
" specificity = tn / (tn + fp)\n",
" return specificity\n",
"\n",
"def sensitivity_score(df, y_true_col_names, y_pred_col_names):\n",
" tp = snow_sum((col(y_true_col_names) == 1) & (col(y_pred_col_names) == 1)).alias(\"TP\")\n",
" fn = snow_sum((col(y_true_col_names) == 1) & (col(y_pred_col_names) == 0)).alias(\"FN\")\n",
" sensitivity = tp / (tp + fn)\n",
" return sensitivity\n",
"\n",
"specificity = specificity_score(result, y_true_col_names=\"SURVIVED\", y_pred_col_names=\"PRED_SURVIVED\")\n",
"sensitivity = sensitivity_score(result, y_true_col_names=\"SURVIVED\", y_pred_col_names=\"PRED_SURVIVED\")\n",
"\n",
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"accuracy = accuracy_score(\n",
" df=result, y_true_col_names=\"SURVIVED\", y_pred_col_names=\"PRED_SURVIVED\"\n",
Expand All @@ -178,11 +201,22 @@
" df=result, y_true_col_names=\"SURVIVED\", y_pred_col_names=\"PRED_SURVIVED\"\n",
")\n",
"\n",
"print(f\"Accuracy: {accuracy}, Precision: {precision}, Recall: {recall}\")"
],
"cell_type": "code",
"outputs": [],
"execution_count": null
"specificity = specificity_score(\n",
" df=result, y_true_col_names=\"SURVIVED\", y_pred_col_names=\"PRED_SURVIVED\"\n",
")\n",
"\n",
"sensitivity = sensitivity_score(\n",
" df=result, y_true_col_names=\"SURVIVED\", y_pred_col_names=\"PRED_SURVIVED\"\n",
")\n",
"\n",
]
}
]
],
"metadata": {
"language_info": {
"name": "python"
}
},
"nbformat": 4,
"nbformat_minor": 2
}