app.py

from transformers import AutoModelForSequenceClassification, AutoTokenizer
import torch
import streamlit as st
import pandas as pd

model1 = AutoModelForSequenceClassification.from_pretrained(
    "rexarski/bert-base-climate-fever-fixed"
)
tokenizer1 = AutoTokenizer.from_pretrained(
    "rexarski/bert-base-climate-fever-fixed"
)

label_mapping1 = ["SUPPORTS", "REFUTES", "NOT_ENOUGH_INFO"]

model2 = AutoModelForSequenceClassification.from_pretrained(
    "rexarski/distilroberta-tcfd-disclosure"
)

tokenizer2 = AutoTokenizer.from_pretrained("distilroberta-base")

label_mapping2 = [
    "Governance a)",
    "Governance b)",
    "Metrics and Targets a)",
    "Metrics and Targets b)",
    "Metrics and Targets c)",
    "Risk Management a)",
    "Risk Management b)",
    "Risk Management c)",
    "Strategy a)",
    "Strategy b)",
    "Strategy c)",
]


def factcheck(text1, text2):
    features = tokenizer1(
        [text1],
        [text2],
        padding="max_length",
        truncation=True,
        return_tensors="pt",
        max_length=512,
    )

    model1.eval()
    with torch.no_grad():
        scores = model1(**features).logits
        labels = [
            label_mapping1[score_max] for score_max in scores.argmax(dim=1)
        ]
    return labels[0]


def tcfd_classify(text):
    features = tokenizer2(
        text,
        padding="max_length",
        truncation=True,
        return_tensors="pt",
        max_length=512,
    )

    model2.eval()
    with torch.no_grad():
        scores = model2(**features).logits
        labels = [
            label_mapping2[score_max] for score_max in scores.argmax(dim=1)
        ]
    return labels[0]


data1 = {
    "example": [
        "Example 1 (there is no relationship between temperature and carbon dioxide emissions by ­humans[...])",
        "Example 2 (The late 1970s marked the end of a 30-year cooling trend.)",
        "Example 3 (Even during a period of long term warming, there are short periods of cooling due to climate variability.)",
        "Example 4 (Humans are too insignificant to affect global climate.)",
    ],
    "claim": [
        "there is no relationship between temperature and carbon dioxide emissions by ­humans[...]",
        "The late 1970s marked the end of a 30-year cooling trend.",
        "Even during a period of long term warming, there are short periods of cooling due to climate variability.",
        "Humans are too insignificant to affect global climate.",
    ],
    "evidence": [
        "Human activities are now causing atmospheric concentrations of greenhouse gases—including carbon dioxide, methane, tropospheric ozone, and nitrous oxide—to rise well above pre-industrial levels ... Increases in greenhouse gases are causing temperatures to rise ...",
        "During the last 20-30 years, world temperature has fallen, irregularly at first but more sharply over the last decade..",
        "El Niño events cause short-term (approximately 1 year in length) spikes in global average surface temperature while La Niña events cause short term cooling.",
        "Human impact on the environment or anthropogenic impact on the environment includes changes to biophysical environments and ecosystems, biodiversity, and natural resources caused directly or indirectly by humans, including global warming, environmental degradation (such as ocean acidification), mass extinction and biodiversity loss, ecological crisis, and ecological collapse.",
    ],
    "label": ["REFUTES", "NOT_ENOUGH_INFO", "SUPPORTS", "REFUTES"],
}

data2 = {
    "example": [
        "Example 1 (Should our products fail to meet energy-efficiency standards and regulations, we will risk losing sales opportunities.)",
        "Example 2 (There are no sentences in the provided excerpts that describe the targets the company uses to manage climate-related risks or opportunities.)",
        "Example 3 (Describe how processes for identifying, assessing, and managing climate-related risks are integrated into the organization’s overall risk management.)",
        "Example 4 (Reporting on such risks and opportunities is provided to.)",
    ],
    "text": [
        "Should our products fail to meet energy-efficiency standards and regulations, we will risk losing sales opportunities.",
        "There are no sentences in the provided excerpts that describe the targets the company uses to manage climate-related risks or opportunities.",
        "Describe how processes for identifying, assessing, and managing climate-related risks are integrated into the organization’s overall risk management.",
        "Reporting on such risks and opportunities is provided to.",
    ],
    "label": [
        "Strategy a)",
        "Metrics and Targets c)",
        "Risk Management c)",
        "Governance a)",
    ],
}


def get_pred_emoji(str1, str2, mode="factcheck"):
    if mode == "factcheck":
        if str1 == str2:
            return "✅"
        else:
            return "❌"
    elif mode == "tcfd":
        if str1 == str2:
            return "✅"
        elif str1.split()[:-1] == str2.split()[:-1]:
            return "🔧"
        else:
            return "❌"


df1 = pd.DataFrame(data1)
df2 = pd.DataFrame(data2)

st.markdown(
    """
# climate-plus demo 🌦️
This is a minimal example of two models we trained for `climate-plus` project:

- [bert-base-climate-fever-fixed](https://huggingface.co/rexarski/bert-base-climate-fever-fixed)
- [distilroberta-tcfd-disclosure](https://huggingface.co/rexarski/distilroberta-tcfd-disclosure)

See the [GitHub repo](https://github.com/rexarski/climate-plus) for more details."
"""
)

st.markdown("## Factchecking 🌤️")

factchecking_demo = st.radio(
    "What examples do you want to see?",
    ("Preloaded examples", "Custom examples"),
    key="factchecking_demo",
)

if factchecking_demo == "Preloaded examples":
    ex1_selected = st.selectbox(
        "Select a climate claim-evidence pair", df1["example"]
    )
    selected_row1 = df1[df1["example"] == ex1_selected]

    ex_claim = selected_row1["claim"].values[0]
    ex_evidence = selected_row1["evidence"].values[0]
    ex_label = selected_row1["label"].values[0]
    if st.button("Run", key="run1"):
        ex_pred = factcheck(
            selected_row1["claim"].values[0],
            selected_row1["evidence"].values[0],
        )
        st.markdown(f"**Claim**: {ex_claim}")
        st.markdown(f"**Evidence**: {ex_evidence}")
        st.markdown(f"**Label**: {ex_label}")
        st.markdown(
            f'**Prediction**: {ex_pred} {get_pred_emoji(ex_label, ex_pred, mode="factcheck")}'
        )
else:
    st.markdown("Or enter your own claim and evidence below:")
    custom_claim = st.text_input(label="Enter your claim.")
    custom_evidence = st.text_input(label="Enter your evidence.")
    if st.button("Run", key="run2"):
        if custom_claim != "" and custom_evidence != "":
            st.markdown(
                f"**Prediction**: {factcheck(custom_claim, custom_evidence)}"
            )

st.markdown("---")

st.markdown("## TCFD disclosure classification ⛈️")

tcfd_demo = st.radio(
    "What examples do you want to see?",
    ("Preloaded examples", "Custom examples"),
    key="tcfd_demo",
)

if tcfd_demo == "Preloaded examples":
    ex2_selected = st.selectbox(
        "Select a TCFD disclosure example", df2["example"]
    )
    selected_row2 = df2[df2["example"] == ex2_selected]

    ex_text = selected_row2["text"].values[0]
    ex_label2 = selected_row2["label"].values[0]
    if st.button("Run", key="run3"):
        ex_pred2 = tcfd_classify(selected_row2["text"].values[0])
        st.markdown(f"**Text**: {ex_text}")
        st.markdown(f"**Label**: {ex_label2}")
        st.markdown(
            f'**Prediction**: {ex_pred2} {get_pred_emoji(ex_label2, ex_pred2, mode="tcfd")}'
        )
else:
    st.markdown(
        "Or enter your own sentence to see if it belongs to any specific TCFD disclosure category:"
    )
    custom_text = st.text_input(label="Enter your text.")
    if st.button("Run", key="run4"):
        if custom_text != "":
            st.markdown(f"**Prediction**: {tcfd_classify(custom_text)}")

st.markdown(
    """
            ---

            🌦️
            """
)