feat: finish logic

Pipfile

@@ -0,0 +1,15 @@
+[[source]]
+url = "https://pypi.org/simple"
+verify_ssl = true
+name = "pypi"
+
+[packages]
+gradio = "*"
+pandas = "*"
+joblib = "*"
+scikit-learn = "*"
+
+[dev-packages]
+
+[requires]
+python_version = "3.9"

app.py

@@ -1,239 +1,312 @@
 import pandas as pd
 import gradio as gr
+import joblib
 
 
 # Inputs for UI
 
-with gr.Blocks(
-    theme=gr.themes.Soft(
-        primary_hue=gr.themes.colors.zinc,
-        neutral_hue=gr.themes.colors.slate,
-    )
-) as demo:
-    age = (
-        gr.Number(
-            label="Age",
-            info="Age of client at admission",
-            minimum=0,
-            maximum=100,
-        ),
-    )
-    race = (
-        gr.Dropdown(
-            [
-                "Alaska Native (Aleut, Eskimo, Indian)",
-                "American Indian (other than Alaska Native)",
-                "Asian or Pacific Islander",
-                "Black or African American",
-                "White",
-                "Asian",
-                "Other single race",
-                "Two or more races",
-                "Native Hawaiian or Other Pacific Islander",
-            ],
-            label="Race",
-        ),
-    )
-    education = (
-        gr.Dropdown(
-            [
-                "Less than one school grade, no schooling, nursery school, or kindergarten to Grade 8",
-                "Grades 9 to 11",
-                "Grade 12 (or GED)",
-                "1-3 years of college, university, or vocational school",
-                "4 years of college, university, BA/BS, some postgraduate study, or more",
-            ],
-            label="Education",
-            info="The highest school grade completed for adults or children not attending school, or current school grade for school-age children (3-17 years old) attending school",
-        ),
-    )
-    marital = (
-        gr.Dropdown(
-            ["Never married", "Now married", "Separated", "Divorced, widowed"],
-            label="Marital Status",
-            info="Client's marital status, compatible with U.S. Census categories",
-        ),
-    )
-    primary_income = (
-        gr.Dropdown(
-            [
-                "Wages/salary",
-                "Public assistance",
-                "Retirement/pension, disability",
-                "Other",
-                "None",
-            ],
-            label="Primary Source of Income/Support",
-            info="Client's principal source of financial support (for children younger than 18 years old, the primary parental source of income/support)",
-        ),
-    )
-    health_insurance = (
-        gr.Dropdown(
-            [
-                "Private insurance, Blue Cross/Blue Shield, HMO",
-                "Medicaid",
-                "Medicare, other (e.g. TRICARE, CHAMPUS)",
-                "None",
-            ],
-            label="Health Insurance",
-            info="Client's health insurance at admission",
-        ),
-    )
-    primary_substance = (
-        gr.Dropdown(
-            [
-                "None",
-                "Alcohol",
-                "Cocaine/crack",
-                "Marijuana/hashish: Includes THC and any other cannabis sativa preparations",
-                "Heroin",
-                "Non-prescription methadone",
-                "Other opiates and synthetics: Includes buprenorphine, butorphanol, codeine, hydrocodone, hydromorphone, meperidine,morphine, opium, oxycodone, pentazocine, propoxyphene, tramadol, and other narcotic analgesics, opiates, or synthetics",
-                "PCP: Phencyclidine",
-                "Hallucinogens: Includes LSD, DMT, mescaline, peyote, psilocybin, STP, and other hallucinogens",
-                "Methamphetamine/speed",
-                "Other amphetamines: Includes amphetamines, MDMA, ‘bath salts’, phenmetrazine, and other amines and related drugs",
-                "Other stimulants: Includes methylphenidate and any other stimulants",
-                "Benzodiazepines: Includes alprazolam, chlordiazepoxide, clonazepam, clorazepate, diazepam, flunitrazepam,flurazepam, halazepam, lorazepam, oxazepam, prazepam, temazepam, triazolam, and other unspecified benzodiazepines",
-                "Other tranquilizers: Includes meprobamate, and other non-benzodiazepine tranquilizers",
-                "Barbiturates: Includes amobarbital, pentobarbital, phenobarbital, secobarbital, etc.",
-                "Other sedatives or hypnotics: Includes chloral hydrate, ethchlorvynol, glutethimide, methaqualone, and othernon-barbiturate sedatives and hypnotics",
-                "Inhalants: Includes aerosols; chloroform, ether, nitrous oxide and other anesthetics; gasoline; glue; nitrites; paint thinnerand other solvents; and other inappropriately inhaled products",
-                "Over-the-counter medications: Includes aspirin, dextromethorphan and other cough syrups, diphenhydramine and otheranti-histamines, ephedrine, sleep aids, and any other legally obtained, non-prescription medication",
-                "Other drugs: Includes diphenylhydantoin/phenytoin, GHB/GBL, ketamine, synthetic cannabinoid 'Spice', carisoprodol(Soma), and other drugs",
-            ],
-            label="Primary Substance Use",
-            info="Client's primary substance use at admission",
-        ),
-    )
-    first_use = (
-        gr.Number(
-            label="Age at First Use of Primary Substance",
-            info="For alcohol use, this is the age of first intoxication",
-            minimum=0,
-            maximum=100,
-        ),
-    )
-    frequency = (
-        gr.Radio(
-            ["No use in the past month", "Some use", "Daily use"],
-            label="Frequency of Primary Substance Use",
-        ),
-    )
-    days_waiting = (
-        gr.Number(
-            label="Days Waiting to Enter Substance Use Treatment",
-            info="Number of days from the first contact or request for substance use treatment service until the client was admitted and the first clinical substance use treatment service was provided",
-            minimum=0,
-        ),
-    )
-    arrests = (
-        gr.Radio(
-            ["None", "Once", "Two or more times"],
-            label="Arrests",
-            info="Number of arrests in the 30 days prior to admission",
-        ),
-    )
-    attendance = (
-        gr.Dropdown(
-            [
-                "No attendance",
-                "1-3 times in the past month",
-                "4-7 times in the past month",
-                "8-30 times in the past month",
-                "Some attendance, frequency is unknown",
-            ],
-            label="Attendance at Substance Use Self-help Groups in Past 30 Days",
-            info="Frequency of attendance at a substance use self-help group in the 30 days prior to the reference date (the date of admission). Includes Alcoholics Anonymous (AA), Narcotics Anonymous (NA), and other self-help/mutual support groups focused on recovery from substance use and dependence",
-        ),
-    )
-    services = (
-        gr.Dropdown(
-            [
-                "Detox, 24-hour, hospital inpatient",
-                "Detox, 24-hour, free-standing residential",
-                "Rehab/residential, hospital (non-detox)",
-                "Rehab/residential, short term (30 days or fewer)",
-                "Rehab/residential, long term (more than 30 days)",
-                "Ambulatory, intensive outpatient",
-                "Ambulatory, non-intensive outpatient",
-                "Ambulatory, detoxification",
-            ],
-            label="Type of Treatment Service/Setting",
-            info="Type of treatment service or treatment setting in which the client is placed at the time of admission or transfer",
-        ),
-    )
-    co_occuring = (
-        gr.Radio(
-            ["Yes", "No"], label="Co-occurring Mental and Substance Use Disorders"
-        ),
-    )
-
-
-def conversion():
+
+def age_conversion(age):
     # Convert inputs to model inputs
-    if age >= 12 or age <= 14:
-        age = 1
-    elif age >= 15 or age <= 17:
-        age = 2
-    elif age >= 18 or age <= 20:
-        age = 3
-    elif age >= 21 or age <= 24:
-        age = 4
-    elif age >= 25 or age <= 29:
-        age = 5
-    elif age >= 30 or age <= 34:
-        age = 6
-    elif age >= 35 or age <= 39:
-        age = 7
-    elif age >= 40 or age <= 44:
-        age = 8
-    elif age >= 45 or age <= 49:
-        age = 9
-    elif age >= 50 or age <= 54:
-        age = 10
-    elif age >= 55 or age <= 64:
-        age = 11
+    if 12 <= age <= 14:
+        age_range = 1
+    elif 15 <= age <= 17:
+        age_range = 2
+    elif 18 <= age <= 20:
+        age_range = 3
+    elif 21 <= age <= 24:
+        age_range = 4
+    elif 25 <= age <= 29:
+        age_range = 5
+    elif 30 <= age <= 34:
+        age_range = 6
+    elif 35 <= age <= 39:
+        age_range = 7
+    elif 40 <= age <= 44:
+        age_range = 8
+    elif 45 <= age <= 49:
+        age_range = 9
+    elif 50 <= age <= 54:
+        age_range = 10
+    elif 55 <= age <= 64:
+        age_range = 11
     else:
-        age = 12
+        age_range = 12
+    return age_range
 
-    race = race.index() + 1
-    education = education.index() + 1
-    marital = marital.index() + 1
-    primary_income = primary_income.index() + 1
-    health_insurance = health_insurance.index() + 1
-    primary_substance = primary_substance.index() + 1
-    frequency = frequency(type="index") + 1
 
+def days_waiting_conversion(days_waiting):
     if days_waiting == 0:
-        days_waiting = 0
-    elif days_waiting >= 1 or days_waiting <= 7:
-        days_waiting = 1
-    elif days_waiting >= 8 or days_waiting <= 14:
-        days_waiting = 2
-    elif days_waiting >= 15 or days_waiting <= 30:
-        days_waiting = 3
+        days_waiting_range = 0
+    elif 1 <= days_waiting <= 7:
+        days_waiting_range = 1
+    elif 8 <= days_waiting <= 14:
+        days_waiting_range = 2
+    elif 15 <= days_waiting <= 30:
+        days_waiting_range = 3
     else:
-        days_waiting = 4
+        days_waiting_range = 4
+
+    return days_waiting_range
+
 
-    arrests = arrests(type="index")
-    attendance = attendance.index() + 1
-    services = services.index() + 1
+def first_use_conversion(first_use):
 
     if first_use <= 11:
-        first_use = 1
-    elif first_use >= 12 or first_use <= 14:
-        first_use = 2
-    elif first_use >= 15 or first_use <= 17:
-        first_use = 3
-    elif first_use >= 18 or first_use <= 20:
-        first_use = 4
-    elif first_use >= 21 or first_use <= 24:
-        first_use = 5
-    elif first_use >= 25 or first_use <= 29:
-        first_use = 6
+        first_use_range = 1
+    elif 12 <= first_use <= 14:
+        first_use_range = 2
+    elif 15 <= first_use <= 17:
+        first_use_range = 3
+    elif 18 <= first_use <= 20:
+        first_use_range = 4
+    elif 21 <= first_use <= 24:
+        first_use_range = 5
+    elif 25 <= first_use <= 29:
+        first_use_range = 6
     else:
-        first_use = 7
+        first_use_range = 7
+    return first_use_range
+
+
+def load_model():
+    # Load the model
+    rf = joblib.load("random_forest_model.joblib")
+    return rf
+
+
+MODEL = load_model()
+
+
+def call_the_model(
+    age,
+    arrests,
+    substance,
+    marital,
+    education,
+    primary_income,
+    days_waiting,
+    frequency,
+    self_help,
+    health_insurance,
+    first_use,
+    race,
+    services,
+    co_occuring,
+):
+    # create a dataframe
+    data = {
+        "AGE": [age_conversion(age)],
+        "ARRESTS": [arrests],
+        "SUB1": [substance + 1],
+        "MARSTAT": [marital + 1],
+        "EDUC": [education + 1],
+        "PRIMINC": [primary_income + 1],
+        "DAYWAIT": [days_waiting_conversion(days_waiting)],
+        "FREQ1": [frequency + 1],
+        "FREQ_ATND_SELF_HELP": [self_help + 1],
+        "HLTHINS": [health_insurance + 1],
+        "FRSTUSE1": [first_use_conversion(first_use)],
+        "RACE": [race + 1],
+        "SERVICES": [services + 1],
+        "PSYPROB": [co_occuring],
+    }
+    input_df = pd.DataFrame(data)
+
+    # print(df)
+    # res = MODEL.predict(input_df)
+    prob = MODEL.predict_proba(input_df)
+    no, yes = prob[0]
+    if no > yes:
+        return f"The client is {round(no * 100)}% likely to relapse."
+    else:
+        return f"The client is {round(yes * 100)}% likely to complete the treatment program."
+
+
+def inference(age: gr.Number, race: gr.Dropdown):
+    pass
+
+
+with gr.Blocks(
+    theme=gr.themes.Soft(
+        primary_hue=gr.themes.colors.zinc,
+        neutral_hue=gr.themes.colors.slate,
+    )
+) as demo:
+    age = gr.Number(
+        label="Age", info="Age of client at admission", minimum=0, maximum=100, value=25
+    )
+    race = gr.Dropdown(
+        [
+            "Alaska Native (Aleut, Eskimo, Indian)",
+            "American Indian (other than Alaska Native)",
+            "Asian or Pacific Islander",
+            "Black or African American",
+            "White",
+            "Asian",
+            "Other single race",
+            "Two or more races",
+            "Native Hawaiian or Other Pacific Islander",
+        ],
+        label="Race",
+        type="index",
+        value=0,
+    )
+    education = gr.Dropdown(
+        [
+            "Less than one school grade, no schooling, nursery school, or kindergarten to Grade 8",
+            "Grades 9 to 11",
+            "Grade 12 (or GED)",
+            "1-3 years of college, university, or vocational school",
+            "4 years of college, university, BA/BS, some postgraduate study, or more",
+        ],
+        label="Education",
+        info="The highest school grade completed for adults or children not attending school, or current school grade for school-age children (3-17 years old) attending school",
+        type="index",
+        value=0,
+    )
+    marital = gr.Dropdown(
+        ["Never married", "Now married", "Separated", "Divorced, widowed"],
+        label="Marital Status",
+        info="Client's marital status, compatible with U.S. Census categories",
+        type="index",
+        value=0,
+    )
+    primary_income = gr.Dropdown(
+        [
+            "Wages/salary",
+            "Public assistance",
+            "Retirement/pension, disability",
+            "Other",
+            "None",
+        ],
+        label="Primary Source of Income/Support",
+        info="Client's principal source of financial support (for children younger than 18 years old, the primary parental source of income/support)",
+        type="index",
+        value=0,
+    )
+    health_insurance = gr.Dropdown(
+        [
+            "Private insurance, Blue Cross/Blue Shield, HMO",
+            "Medicaid",
+            "Medicare, other (e.g. TRICARE, CHAMPUS)",
+            "None",
+        ],
+        label="Health Insurance",
+        info="Client's health insurance at admission",
+        type="index",
+        value=0,
+    )
+    primary_substance = gr.Dropdown(
+        [
+            "None",
+            "Alcohol",
+            "Cocaine/crack",
+            "Marijuana/hashish: Includes THC and any other cannabis sativa preparations",
+            "Heroin",
+            "Non-prescription methadone",
+            "Other opiates and synthetics: Includes buprenorphine, butorphanol, codeine, hydrocodone, hydromorphone, meperidine,morphine, opium, oxycodone, pentazocine, propoxyphene, tramadol, and other narcotic analgesics, opiates, or synthetics",
+            "PCP: Phencyclidine",
+            "Hallucinogens: Includes LSD, DMT, mescaline, peyote, psilocybin, STP, and other hallucinogens",
+            "Methamphetamine/speed",
+            "Other amphetamines: Includes amphetamines, MDMA, ‘bath salts’, phenmetrazine, and other amines and related drugs",
+            "Other stimulants: Includes methylphenidate and any other stimulants",
+            "Benzodiazepines: Includes alprazolam, chlordiazepoxide, clonazepam, clorazepate, diazepam, flunitrazepam,flurazepam, halazepam, lorazepam, oxazepam, prazepam, temazepam, triazolam, and other unspecified benzodiazepines",
+            "Other tranquilizers: Includes meprobamate, and other non-benzodiazepine tranquilizers",
+            "Barbiturates: Includes amobarbital, pentobarbital, phenobarbital, secobarbital, etc.",
+            "Other sedatives or hypnotics: Includes chloral hydrate, ethchlorvynol, glutethimide, methaqualone, and othernon-barbiturate sedatives and hypnotics",
+            "Inhalants: Includes aerosols; chloroform, ether, nitrous oxide and other anesthetics; gasoline; glue; nitrites; paint thinnerand other solvents; and other inappropriately inhaled products",
+            "Over-the-counter medications: Includes aspirin, dextromethorphan and other cough syrups, diphenhydramine and otheranti-histamines, ephedrine, sleep aids, and any other legally obtained, non-prescription medication",
+            "Other drugs: Includes diphenylhydantoin/phenytoin, GHB/GBL, ketamine, synthetic cannabinoid 'Spice', carisoprodol(Soma), and other drugs",
+        ],
+        label="Primary Substance Use",
+        info="Client's primary substance use at admission",
+        type="index",
+        value=0,
+    )
+    first_use = gr.Number(
+        label="Age at First Use of Primary Substance",
+        info="For alcohol use, this is the age of first intoxication",
+        minimum=0,
+        maximum=100,
+        value=25,
+    )
+    frequency = gr.Radio(
+        ["No use in the past month", "Some use", "Daily use"],
+        label="Frequency of Primary Substance Use",
+        type="index",
+        value="No use in the past month",
+    )
+    days_waiting = gr.Number(
+        label="Days Waiting to Enter Substance Use Treatment",
+        info="Number of days from the first contact or request for substance use treatment service until the client was admitted and the first clinical substance use treatment service was provided",
+        minimum=0,
+    )
+    arrests = gr.Radio(
+        ["None", "Once", "Two or more times"],
+        label="Arrests",
+        info="Number of arrests in the 30 days prior to admission",
+        type="index",
+        value="None",
+    )
+    attendance = gr.Dropdown(
+        [
+            "No attendance",
+            "1-3 times in the past month",
+            "4-7 times in the past month",
+            "8-30 times in the past month",
+            "Some attendance, frequency is unknown",
+        ],
+        label="Attendance at Substance Use Self-help Groups in Past 30 Days",
+        info="Frequency of attendance at a substance use self-help group in the 30 days prior to the reference date (the date of admission). Includes Alcoholics Anonymous (AA), Narcotics Anonymous (NA), and other self-help/mutual support groups focused on recovery from substance use and dependence",
+        type="index",
+        value=0,
+    )
+    services = gr.Dropdown(
+        [
+            "Detox, 24-hour, hospital inpatient",
+            "Detox, 24-hour, free-standing residential",
+            "Rehab/residential, hospital (non-detox)",
+            "Rehab/residential, short term (30 days or fewer)",
+            "Rehab/residential, long term (more than 30 days)",
+            "Ambulatory, intensive outpatient",
+            "Ambulatory, non-intensive outpatient",
+            "Ambulatory, detoxification",
+        ],
+        label="Type of Treatment Service/Setting",
+        info="Type of treatment service or treatment setting in which the client is placed at the time of admission or transfer",
+        type="index",
+        value=0,
+    )
+
+    co_occuring = gr.Checkbox(
+        value=False, label="Co-occurring Mental and Substance Use Disorders"
+    )
+
+    submit_btn = gr.Button("SUBMIT", variant="primary", size="large")
+
+    result = gr.Label(label="Result")
+
+    submit_btn.click(
+        call_the_model,
+        [
+            age,
+            arrests,
+            primary_substance,
+            marital,
+            education,
+            primary_income,
+            days_waiting,
+            frequency,
+            attendance,
+            health_insurance,
+            first_use,
+            race,
+            services,
+            co_occuring,
+        ],
+        [result],
+    )
 
 
 if __name__ == "__main__":

random_forest_model.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eaf166a1d1be8c79a0ebbb4b090bff7562e101e2ba39d03889f73769feb45da
+size 513118921

requirements.txt

@@ -0,0 +1,72 @@
+-i https://pypi.org/simple
+aiofiles==23.2.1; python_version >= '3.7'
+altair==5.2.0; python_version >= '3.8'
+annotated-types==0.6.0; python_version >= '3.8'
+anyio==4.2.0; python_version >= '3.8'
+attrs==23.2.0; python_version >= '3.7'
+certifi==2024.2.2; python_version >= '3.6'
+charset-normalizer==3.3.2; python_full_version >= '3.7.0'
+click==8.1.7; python_version >= '3.7'
+colorama==0.4.6
+contourpy==1.2.0; python_version >= '3.9'
+cycler==0.12.1; python_version >= '3.8'
+exceptiongroup==1.2.0; python_version < '3.11'
+fastapi==0.109.2; python_version >= '3.8'
+ffmpy==0.3.1
+filelock==3.13.1; python_version >= '3.8'
+fonttools==4.48.1; python_version >= '3.8'
+fsspec==2024.2.0; python_version >= '3.8'
+gradio==4.18.0; python_version >= '3.8'
+gradio-client==0.10.0; python_version >= '3.8'
+h11==0.14.0; python_version >= '3.7'
+httpcore==1.0.2; python_version >= '3.8'
+httpx==0.26.0; python_version >= '3.8'
+huggingface-hub==0.20.3; python_full_version >= '3.8.0'
+idna==3.6; python_version >= '3.5'
+importlib-resources==6.1.1; python_version >= '3.8'
+jinja2==3.1.3; python_version >= '3.7'
+joblib==1.3.2; python_version >= '3.7'
+jsonschema==4.21.1; python_version >= '3.8'
+jsonschema-specifications==2023.12.1; python_version >= '3.8'
+kiwisolver==1.4.5; python_version >= '3.7'
+markdown-it-py==3.0.0; python_version >= '3.8'
+markupsafe==2.1.5; python_version >= '3.7'
+matplotlib==3.8.2; python_version >= '3.9'
+mdurl==0.1.2; python_version >= '3.7'
+numpy==1.26.4; python_version < '3.11'
+orjson==3.9.13; python_version >= '3.8'
+packaging==23.2; python_version >= '3.7'
+pandas==2.2.0; python_version >= '3.9'
+pillow==10.2.0; python_version >= '3.8'
+pydantic==2.6.1; python_version >= '3.8'
+pydantic-core==2.16.2; python_version >= '3.8'
+pydub==0.25.1
+pygments==2.17.2; python_version >= '3.7'
+pyparsing==3.1.1; python_full_version >= '3.6.8'
+python-dateutil==2.8.2; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3'
+python-multipart==0.0.9; python_version >= '3.8'
+pytz==2024.1
+pyyaml==6.0.1; python_version >= '3.6'
+referencing==0.33.0; python_version >= '3.8'
+requests==2.31.0; python_version >= '3.7'
+rich==13.7.0
+rpds-py==0.17.1; python_version >= '3.8'
+ruff==0.2.1; python_version >= '3.7'
+scikit-learn==1.4.0; python_version >= '3.9'
+scipy==1.12.0; python_version >= '3.9'
+semantic-version==2.10.0; python_version >= '2.7'
+shellingham==1.5.4
+six==1.16.0; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3'
+sniffio==1.3.0; python_version >= '3.7'
+starlette==0.36.3; python_version >= '3.8'
+threadpoolctl==3.2.0; python_version >= '3.8'
+tomlkit==0.12.0; python_version >= '3.7'
+toolz==0.12.1; python_version >= '3.7'
+tqdm==4.66.2; python_version >= '3.7'
+typer[all]==0.9.0; python_version >= '3.6'
+typing-extensions==4.9.0; python_version >= '3.8'
+tzdata==2023.4; python_version >= '2'
+urllib3==2.2.0; python_version >= '3.8'
+uvicorn==0.27.1; python_version >= '3.8'
+websockets==11.0.3; python_version >= '3.7'
+zipp==3.17.0; python_version < '3.10'