app.py

import pandas as pd
import streamlit as st

# from annotated_text import annotated_text
from annotated_text.util import get_annotated_html
from streamlit_annotation_tools import text_labeler

from constants import (
    APP_INTRO,
    APP_TITLE,
    EVAL_FUNCTION_INTRO,
    EVAL_FUNCTION_PROPERTIES,
    NER_TASK_EXPLAINER,
    PREDICTION_ADDITION_INSTRUCTION,
)
from evaluation_metrics import EVALUATION_METRICS
from predefined_example import EXAMPLES
from span_dataclass_converters import (
    get_highlight_spans_from_ner_spans,
    get_ner_spans_from_annotations,
)


@st.cache_resource
def get_examples_attributes(selected_example):
    "Return example attributes so that they are not refreshed on every interaction"
    return (
        selected_example.text,
        selected_example.gt_labels,
        selected_example.gt_spans,
        selected_example.predictions,
        selected_example.tags,
    )


if __name__ == "__main__":
    st.set_page_config(layout="wide")
    st.title(APP_TITLE)

    st.write(APP_INTRO)
    explanation_tab, comparision_tab = st.tabs(["📙 Explanation", "⚖️ Comparision"])

    with explanation_tab:
        st.write(EVAL_FUNCTION_INTRO)
        st.image("assets/eval_fnc_viz.png", caption="Evaluation Function Flow")
        st.markdown(EVAL_FUNCTION_PROPERTIES)
        st.markdown(NER_TASK_EXPLAINER)
        st.subheader("Evaluation Metrics")
        metric_names = "\n".join(
            [
                f"{index+1}. " + evaluation_metric.name
                for index, evaluation_metric in enumerate(EVALUATION_METRICS)
            ]
        )
        st.markdown(
            "The different evaluation metrics we have for the NER task are\n"
            "\n"
            f"{metric_names}"
        )

    with comparision_tab:
        # with st.container():
        st.subheader("Ground Truth & Predictions")  # , divider='rainbow')

        selected_example = st.selectbox(
            "Select an example text from the drop down below",
            [example for example in EXAMPLES],
            format_func=lambda ex: ex.text,
        )

        text, gt_labels, gt_spans, predictions, tags = get_examples_attributes(
            selected_example
        )

        # annotated_text(
        #     get_highlight_spans_from_ner_spans(
        #         get_ner_spans_from_annotations(gt_labels), text
        #     )
        # )

        annotated_predictions = [
            get_annotated_html(get_highlight_spans_from_ner_spans(ner_span, text))
            for ner_span in predictions
        ]
        predictions_df = pd.DataFrame(
            {
                # "ID": [f"Prediction_{index}" for index in range(len(predictions))],
                "Prediction": annotated_predictions,
                "ner_spans": predictions,
            },
            index=["Ground Truth"]
            + [f"Prediction_{index}" for index in range(len(predictions) - 1)],
        )

        # st.subheader("Predictions")  # , divider='rainbow')

        with st.expander("Click to Add Predictions"):
            st.subheader("Adding predictions")
            st.markdown(PREDICTION_ADDITION_INSTRUCTION)
            st.write(
                "Note: Only the spans of the selected label name are shown at a given instance. Click on the label to see the corresponding spans. (or view the json below)",
            )
            labels = text_labeler(text, gt_labels)
            st.json(labels, expanded=False)

            # if st.button("Add Prediction"):
            # labels = text_labeler(text)
            if st.button("Add!"):
                spans = get_ner_spans_from_annotations(labels)
                spans = sorted(spans, key=lambda span: span["start"])
                predictions.append(spans)
                annotated_predictions.append(
                    get_annotated_html(get_highlight_spans_from_ner_spans(spans, text))
                )
                predictions_df = pd.DataFrame(
                    {
                        # "ID": [f"Prediction_{index}" for index in range(len(predictions))],
                        "Prediction": annotated_predictions,
                        "ner_spans": predictions,
                    },
                    index=["Ground Truth"]
                    + [f"Prediction_{index}" for index in range(len(predictions) - 1)],
                )
                print("added")

        highlighted_predictions_df = predictions_df[["Prediction"]]
        st.write(
            highlighted_predictions_df.to_html(escape=False), unsafe_allow_html=True
        )
        st.divider()

        ### EVALUATION METRICS COMPARISION ###

        st.subheader("Evaluation Metrics Comparision")  # , divider='rainbow')
        metric_names = "\n".join(
            ["- " + evaluation_metric.name for evaluation_metric in EVALUATION_METRICS]
        )
        st.markdown(
            "The different evaluation metrics we have for the NER task are\n"
            f"{metric_names}"
        )

        with st.expander("View Predictions Details"):
            st.write(predictions_df.to_html(escape=False), unsafe_allow_html=True)

        if st.button("Get Metrics!"):
            for evaluation_metric in EVALUATION_METRICS:
                predictions_df[evaluation_metric.name] = predictions_df.ner_spans.apply(
                    lambda ner_spans: evaluation_metric.get_evaluation_metric(
                        # metric_type=evaluation_metric_type,
                        gt_ner_span=gt_spans,
                        pred_ner_span=ner_spans,
                        text=text,
                        tags=tags,
                    )
                )

            metrics_df = predictions_df.drop(["ner_spans"], axis=1)

            st.write(metrics_df.to_html(escape=False), unsafe_allow_html=True)