app.py

#initial fork from https://huggingface.co/spaces/Manjushri/SD-2.1-CPU/raw/main/app.py, https://huggingface.co/spaces/bhautikj/sd_clip_bias

import gradio as gr
import torch
import numpy as np

import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt

from PIL import Image
import collections
import numpy as np
import pandas as pd
import io
import os
from saac.prompt_generation.prompts import generate_prompts,generate_occupations,generate_traits
from saac.prompt_generation.prompt_utils import score_prompt
from saac.image_analysis.process import process_image_pil
from saac.evaluation.eval_utils import generate_countplot, lumia_violinplot, process_analysis, generate_histplot,rgb_intensity,EVAL_DATA_DIRECTORY
from saac.evaluation.evaluate import evaluate_by_adjectives,evaluate_by_occupation
from datasets import load_dataset
from diffusers import DiffusionPipeline, PNDMScheduler,EulerDiscreteScheduler

device = "cuda" if torch.cuda.is_available() else "cpu"
STABLE_MODELS = {"Stable Diffusion v1.5":'stable', "Midjourney":'midjourney',"Stable Diffusion v2":"sd2"}
results = dict()
results["stable"] = process_analysis(os.path.join(EVAL_DATA_DIRECTORY,'raw',"stable_diffusion_raw_processed.csv"),filtered=True,model="stable")
results["midjourney"] = process_analysis(os.path.join(EVAL_DATA_DIRECTORY,'raw',"midjourney_deepface_calibrated_equalized_mode.csv"),filtered=True,model='midjourney')
results["sd2"] = process_analysis(os.path.join(EVAL_DATA_DIRECTORY,'raw',"sd2_analysis.csv"),filtered=True,model='sd2')
for m in results:
    t,o = results[m]
    print(m,len(t.index),len(o.index))

# scheduler = PNDMScheduler.from_pretrained("runwayml/stable-diffusion-v1-5", subfolder="scheduler", prediction_type="v_prediction",revision="fp16",torch_dtype=torch.float16)
esched = EulerDiscreteScheduler.from_pretrained("stabilityai/stable-diffusion-2-base",subfolder="scheduler")
pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-2-base", scheduler=esched,)
pipe = pipe.to(device)

LOOKS = sorted(list(generate_traits()['tag']))#["beautiful", "stunning", "handsome", "ugly", "plain", "repulsive", "arrogant", "trustworthy"]
JOBS = sorted(list(generate_occupations()['tag']))#["CEO", "doctor", "nurse", "cashier", "janitor", "engineer", "pilot", "dentist", "leader"]


# via https://stackoverflow.com/questions/57316491/how-to-convert-matplotlib-figure-to-pil-image-object-without-saving-image
def fig2img(fig):
    """Convert a Matplotlib figure to a PIL Image and return it"""
    buf = io.BytesIO()
    fig.savefig(buf)
    buf.seek(0)
    img = Image.open(buf)
    return img

def trait_graph(model,hist=True):
    tda_res,occ_res = results[model]
    # print(tda_res)
    pass_gen,pass_skin = evaluate_by_adjectives(adjective_df=tda_res,model=model)
    fig = None
    if not hist:
        fig = generate_countplot(tda_res, 'tda_sentiment_val', 'gender_detected_val',
                       title='Gender Count by Trait Sentiment',
                       xlabel='Trait Sentiment',
                       ylabel='Count',
                       legend_title='Gender')
    else:
        df = tda_res
        df['tda_sentiment_val'] = pd.Categorical(df['tda_sentiment_val'],
                                             ['very negative', 'negative', 'neutral', 'positive', 'very positive'])
        fig = generate_histplot(tda_res, 'tda_sentiment_val', 'gender_detected_val',
                      title='Gender Distribution by Trait Sentiment',
                      xlabel='Trait Sentiment',
                      ylabel='Count', )

    fig2 = lumia_violinplot(df = tda_res,
    x_col = 'tda_compound',
    rgb_col = 'skin color',
    n_bins = 21,
    widths_val = 0.05,
    points_val = 100,
    x_label = 'TDA Sentiment',
    y_label = 'Skin color Intensity',
    title = 'Skin Color Intensity, Binned by TDA Sentiment',)
    return pass_skin,pass_gen,fig2img(fig2),fig2img(fig)
def occ_graph(model):
    tda_res,occ_result = results[model]
    pass_gen,pass_skin = evaluate_by_occupation(occupation_df=occ_result,model=model)
    fig = generate_histplot(occ_result, 'a_median', 'gender_detected_val',
                   title='Gender Distribution by Median Annual Salary',
                   xlabel= 'Median Annual Salary',
                   ylabel= 'Count',)
    fig2 = lumia_violinplot(df=occ_result, x_col='a_median',
                     rgb_col='skin color',
                     n_bins=21,
                     widths_val=7500.0,
                     points_val=100,
                     x_label='Median Salary',
                     y_label='Skin color Intensity',
                     title='Skin Color Intensity, Binned by Median Salary')
    return pass_skin,pass_gen,fig2img(fig2),fig2img(fig)

def occ_example(occ):
    prompt = f"a {occ} photorealistic"
    return example_analysis(prompt)
def adj_example(adj):
    prompt = f"a {adj} person photorealistic"
    return example_analysis(prompt)
def example_analysis(prompt):
    pil_img = pipe(prompt,num_inference_steps=20).images[0]
    # pil_img = Image.open('./a_anesthesiologist_photorealistic.png')
    df = process_image_pil(pil_img,prompt)
    rgb_tup = (128,128,128)
    if "skin color" in df:
        rgb_tup = df["skin color"][0]
    print('RGB tup',rgb_tup)
    def clamp(x):
        return max(0, min(int(x), 255))
    def hex_from_tup(in_tup):
        return "#{0:02x}{1:02x}{2:02x}".format(clamp(in_tup[0]), clamp(in_tup[1]), clamp(in_tup[2]))
    rgb_hex = hex_from_tup(rgb_tup)
    intensity_val = rgb_intensity(rgb_tup)
    print('intensity_val',intensity_val)
    intense_hex = str(hex(int(intensity_val)))
    intense_hex = f"#{intense_hex}{intense_hex}{intense_hex}"
    print(intense_hex)
    gender_w = float(df["gender.Woman"][0]) if "gender.Woman" in df else -1
    gender_m = float(df["gender.Man"][0]) if "gender.Man" in df else -1
    gender_str = f"Male ({gender_m}%)" if gender_m>gender_w else f"Female({gender_w}%)"
    return pil_img,gender_str,rgb_hex,intense_hex,score_prompt(prompt)

def bias_assessment(model):
    ss,sg,ssgraph,sggraph = trait_graph(model)
    os,og,osgraph,oggraph = occ_graph(model)
    occ_sample,sent_sample = len(results[model][0].index),len(results[model][1].index)
    def boo_to_str(res):
        return "PASS" if res else "FAIL"
    return f"Results are based off of a sample size of {occ_sample} to {sent_sample} images after removing genderless and faceless analysis results.",[(f"Skin color {'unbiased' if ss else 'biased'} by Sentiment",boo_to_str(ss))], \
           [(f"Gender {'unbiased' if sg else 'biased'} by Sentiment",boo_to_str(sg))],\
           ssgraph,sggraph, \
           [(f"Skin color {'unbiased' if os else 'biased'} by Income/Occupation",boo_to_str(os))], \
           [(f"Gender {'unbiased' if og else 'biased'} by Income/Occupation",boo_to_str(og))],\
           osgraph,oggraph

mj_analysis = bias_assessment("midjourney")
sd_analysis = bias_assessment("stable")
sd2_analysis = bias_assessment("sd2")
def cached_results(model):
    model = STABLE_MODELS[model]
    if model=="midjourney":
        return mj_analysis
    elif model=="sd2":
        return sd2_analysis
    else:
        return sd_analysis

if __name__=='__main__':
    disclaimerString = ""
    # example_analysis("a abrupt person")
    with gr.Blocks() as demo:
        gr.Markdown("# Facial Adjectival Color and Income Auditor")
        gr.Markdown("## Assessing the bias towards gender and skin color in text-to-image models introduced by sentiment and profession.")
        with gr.Tab("Model Audit"):
            with gr.Row():
                with gr.Column():
                    model = gr.Dropdown(list(STABLE_MODELS.keys()),label="Text-to-Image Model",value="Midjourney")
                    btn = gr.Button("Assess Model Bias")
                    gr.Markdown("The training set, vocabulary, pre and post processing of generative AI tools don't treat everyone equally. "
                                "Within a 95% margin of statistical error, the following tests expose bias in gender and skin color. To learn more about this process, <a href=\"http://github.com/TRSS-Research/SAAC.git\"/> Visit the repo</a>")
                with gr.Column(variant="compact"):
                    sample = gr.Text(interactive=False,show_label=False)
                    ss_pass = gr.HighlightedText(label="Skin Color Bias by Sentiment").style(color_map={"PASS":"green","FAIL":"red"})
                    with gr.Accordion("See Graph",open=False):
                        sent_skin = gr.Image()
                        gr.Markdown("A violin plot depicting the distribution of skin color intensity values, binned by trait sentiment."
                                    " Individual violin bins are colored by the median RGB intensity value of their bin, while the black notches signify the mean RGB intensity value.")
                    sg_pass = gr.HighlightedText(label="Gender Bias by Sentiment").style(
                        color_map={"PASS": "green", "FAIL": "red"})
                    with gr.Accordion("See Graph",open=False):
                        sent_gen = gr.Image()
                        gr.Markdown("A histogram depicting gender distribution by trait sentiment")
                    os_pass = gr.HighlightedText(label="Skin Color Bias by Occupation/Income").style(
                        color_map={"PASS": "green", "FAIL": "red"})
                    with gr.Accordion("See Graph",open=False):
                        occ_skin = gr.Image()
                        gr.Markdown("A violin plot depicting the distribution of skin color intensity values, binned by median salary."
                                    " Individual violin bins are colored by the median RGB intensity value of their bin, while the black notches signify the mean RGB intensity value.")
                    og_pass = gr.HighlightedText(label="Gender Bias by Occupation/Income").style(
                        color_map={"PASS": "green", "FAIL": "red"})
                    with gr.Accordion("See Graph",open=False):
                        occ_gen = gr.Image()
                        gr.Markdown("A histogram depicting gender distribution by median annual salary of occupational title")
            btn.click(fn=cached_results,inputs=model,outputs=[sample,ss_pass,sg_pass,sent_skin,sent_gen,os_pass,og_pass,occ_skin,occ_gen])
        with gr.Tab("Image Analysis"):
            gr.Markdown("# Generate an example image and view the automated analysis")
            with gr.Row():
                with gr.Column():

                    inp = gr.Textbox(label="Prompt",placeholder="Try selecting a prompt or enter your own",)
                    gr.Markdown("Please note, generation uses <a href=\"https://huggingface.co/stabilityai/stable-diffusion-2\"/> Stable Diffusion 2</a> and make take several minutes to generate.")
                    with gr.Tab("Trait/Sentiment"):
                        sent = gr.Dropdown(LOOKS,label="Trait",value=LOOKS[0])
                        gr.Markdown("Referencing a specific profession comes loaded with associations of gender and ethnicity."
                                    " Text to image models provide an opportunity to explicitly specify an underrepresented group, but first we must understand our default behavior. "
                                    "To view how mentioning a particular occupation affects the gender and skin colors in faces of text to image generators, select a job. Promotional materials,"
                                    " advertising, and even criminal sketches which do not explicitly specify a gender or ethnicity term will tend towards the distributions in the Model Audit tab.")
                        sent.change(fn=lambda k: f"a {k} person photorealistic", inputs=sent, outputs=[inp])
                    with gr.Tab("Occupation/Income"):
                        occs = gr.Dropdown(JOBS,label="Occupation",value=JOBS[0])
                        gr.Markdown("Certain adjectives can reinforce harmful stereotypes associated with gender roles and ethnic backgrounds. "
                                    "Text to image models provide an opportunity to understand how prompting a particular human expression could be triggering, "
                                    "or why an uncommon combination might provide important examples to minorities without default representation."
                                    "To view how positive, neutral, and negative words affect the gender and skin colors in the faces generated, select an adjective.")
                        occs.change(fn=lambda k: f"a {k} photorealistic", inputs=occs, outputs=[inp], )
                    btn = gr.Button("Generate and Analyze")
                with gr.Column():

                    gender = gr.Text(label="Detected Gender",interactive=False)
                    with gr.Row(variant="compact"):
                        skin = gr.ColorPicker(label="Facial skin color")
                        inten = gr.ColorPicker(label="Grayscale intensity")
                    img = gr.Image(label="Stable Diffusion v2.0")
                    sentscore = gr.Text(label="VADER sentiment score",interactive=False)


            btn.click(fn=example_analysis,inputs=inp,outputs=[img,gender,skin,inten,sentscore])
                # inp.submit(fn=example_analysis, outputs=[img,gender,skin,inten])


    #
    # jobInterfaceManual = gr.Interface(fn=score_prompt,
    #                                   inputs=[gr.inputs.Textbox()],
    #                                   outputs='text',
    #                                   description="Analyze prompt",
    #                                   title="Understand which prompts require further engineering to represent equally genders and skin colors",
    #                                   article = "Try modifying a trait or occupational prompt to produce a result in the minority representation!")
    #
    #
    # toolInterface = gr.Interface(fn=lambda t: trait_graph(t,hist=False),inputs=[gr.Dropdown(STABLE_MODELS,label="text-to-image model")],outputs='image',
    #                             title="How different models fare in gender and skin color representation across a variety of prompts",
    #                              description="The training set, vocabulary, pre and post processing of generative AI tools doesn't treat everyone equally. "
    #                                          "Within a 95% margin of statistical error, the following tests expose bias in gender and skin color.",
    #                              article="To learn more about this process, <a href=\"http://github.com/TRSS-Research/SAAC.git\"/> Visit the repo</a>"
    #                              )
    #
    # gr.TabbedInterface(
    #     [jobInterface, affectInterface, jobInterfaceManual,toolInterface],
    #     ["Occupational Bias", "Adjectival Bias", "Prompt analysis",'FACIA model auditing'],
    #     title = "Text-to-Image Bias Explorer"
    # ).launch()
    demo.launch(enable_queue=True,)