Initial space commit

.idea/.gitignore

@@ -0,0 +1,8 @@
+# Default ignored files
+/shelf/
+/workspace.xml
+# Editor-based HTTP Client requests
+/httpRequests/
+# Datasource local storage ignored files
+/dataSources/
+/dataSources.local.xml

.idea/inspectionProfiles/profiles_settings.xml

@@ -0,0 +1,6 @@
+<component name="InspectionProjectProfileManager">
+  <settings>
+    <option name="USE_PROJECT_PROFILE" value="false" />
+    <version value="1.0" />
+  </settings>
+</component>

.idea/misc.xml

@@ -0,0 +1,4 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectRootManager" version="2" project-jdk-name="Python 3.10 (zeroshot)" project-jdk-type="Python SDK" />
+</project>

.idea/modules.xml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectModuleManager">
+    <modules>
+      <module fileurl="file://$PROJECT_DIR$/.idea/zshot.iml" filepath="$PROJECT_DIR$/.idea/zshot.iml" />
+    </modules>
+  </component>
+</project>

.idea/vcs.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="VcsDirectoryMappings">
+    <mapping directory="$PROJECT_DIR$" vcs="Git" />
+  </component>
+</project>

.idea/zshot.iml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<module type="PYTHON_MODULE" version="4">
+  <component name="NewModuleRootManager">
+    <content url="file://$MODULE_DIR$" />
+    <orderEntry type="jdk" jdkName="Python 3.10 (zeroshot)" jdkType="Python SDK" />
+    <orderEntry type="sourceFolder" forTests="false" />
+  </component>
+</module>

app.py

@@ -0,0 +1,171 @@
+import streamlit as st
+
+st.set_page_config(
+    layout="wide",  # Can be "centered" or "wide". In the future also "dashboard", etc.
+    initial_sidebar_state="auto",  # Can be "auto", "expanded", "collapsed"
+    page_title='ZShot',  # String or None. Strings get appended with "• Streamlit".
+    page_icon='./logo_zshot.png',  # String, anything supported by st.image, or None.
+)
+
+import os
+import sys
+
+import spacy
+from zshot.linker import LinkerSMXM, LinkerTARS, LinkerRegen
+from zshot.utils.data_models import Entity
+from zshot.mentions_extractor import MentionsExtractorSpacy
+from zshot.mentions_extractor.utils import ExtractorType
+from zshot import PipelineConfig
+from spacy import displacy
+
+sys.path.append(os.path.abspath('./'))
+import streamlit_apps_config as config
+
+## Marking down NER Style
+st.markdown(config.STYLE_CONFIG, unsafe_allow_html=True)
+
+########## To Remove the Main Menu Hamburger ########
+
+hide_menu_style = """
+        <style>
+        #MainMenu {visibility: hidden;}
+        </style>
+        """
+st.markdown(hide_menu_style, unsafe_allow_html=True)
+
+########## Side Bar ########
+
+## loading logo(newer version with href)
+import base64
+
+
+@st.cache(allow_output_mutation=True)
+def get_base64_of_bin_file(bin_file):
+    with open(bin_file, 'rb') as f:
+        data = f.read()
+    return base64.b64encode(data).decode()
+
+
+@st.cache(allow_output_mutation=True)
+def get_img_with_href(local_img_path, target_url, size='big'):
+    img_format = os.path.splitext(local_img_path)[-1].replace('.', '')
+    bin_str = get_base64_of_bin_file(local_img_path)
+    height = '90%' if size == 'big' else '45%'
+    width = '90%' if size == 'big' else '45%'
+    html_code = f'''
+        <a href="{target_url}" style='text-align: center;'>
+            <img height="{height}" width="{width}" style='display: block; margin-left: auto; margin-right: auto;' src="data:image/{img_format};base64,{bin_str}" />
+        </a>'''
+    return html_code
+
+
+logo_html = get_img_with_href('./logo.png', 'https://www.ibm.com/')
+st.sidebar.markdown(logo_html, unsafe_allow_html=True)
+logo_html = get_img_with_href('./logo_zshot.png', 'https://github.com/IBM/zshot', size='small')
+st.sidebar.markdown(logo_html, unsafe_allow_html=True)
+
+# sidebar info
+linkers = ["REGEN", "SMXM", "TARS"]
+st.sidebar.title("Linker to test")
+selected_model = st.sidebar.selectbox("", linkers)
+
+######## Main Page #########
+
+if selected_model == "REGEN":
+    app_title = "GENRE Linker"
+    app_description = "GENRE performs retrieval generating the unique entity name conditioned on the input text using constrained beam search to only generate valid identifiers."
+    st.title(app_title)
+    st.markdown("<h2>" + app_description + "</h2>", unsafe_allow_html=True)
+
+elif selected_model == "SMXM":
+    app_title = "SMXM Linker"
+    app_description = "SMXM model uses the description of the entities to give the model information about the entities."
+    st.title(app_title)
+    st.markdown("<h2>" + app_description + "</h2>", unsafe_allow_html=True)
+
+elif selected_model == "TARS":
+    app_title = "TARS Linker"
+    app_description = "TARS doesn't need the descriptions of the entities, so if you can't provide the descriptions of the entities maybe this is the approach you're looking for."
+    st.title(app_title)
+    st.markdown("<h2>" + app_description + "</h2>", unsafe_allow_html=True)
+
+st.subheader("")
+
+if 'entities' not in st.session_state:
+    st.session_state['entities'] = [
+        Entity(name="company", description="The name of a company"),
+        Entity(name="location", description="A physical location"),
+        Entity(name="chemical compound", description="Any substance composed of identical molecules consisting of atoms of two or more chemical elements.")
+    ]
+
+def add_ent():
+    st.session_state['entities'].append(Entity(name=st.session_state["name"], description=st.session_state["description"]))
+    st.session_state['name'] = ""
+    st.session_state['description'] = ''
+    st.write(st.session_state["name"])
+    st.write(st.session_state["description"])
+
+for i, entity in enumerate(st.session_state['entities']):
+    col1, col2, col3 = st.columns([2, 5, 1])
+    with col1:
+        st.text(entity.name)
+    with col2:
+        st.text(entity.description)
+    with col3:
+        b = st.button('Remove', key=f"ent_{i}")
+        if b:
+            st.session_state['entities'].pop(i)
+            st.experimental_rerun()  # This causes the app to rerun
+
+with st.form(key="form"):
+    col1, col2, col3 = st.columns([2, 5, 1])
+    with col1:
+        st.text_input("Entity Name", key="name")
+    with col2:
+        st.text_input("Entity Description", key="description")
+    with col3:
+        st.form_submit_button('Add', on_click=add_ent)
+
+st.markdown("________")
+text = st.text_input("Type here your text and press enter to run:",
+                     value="CH2O2 is a chemical compound similar to Acetamide used in International Business "
+                           "Machines Corporation (IBM) to create new materials that act like PAGs.")
+
+def build_pipeline(model_name=selected_model):
+    nlp = spacy.blank('en')
+    mentions_extractor = None
+
+    if model_name == "REGEN":
+        linker = LinkerRegen()
+        nlp = spacy.load('en_core_web_sm')
+        mentions_extractor = MentionsExtractorSpacy(ExtractorType.NER)
+    elif model_name == "TARS":
+        linker = LinkerTARS()
+    elif model_name == "SMXM":
+        linker = LinkerSMXM()
+
+    config = PipelineConfig(
+        entities=st.session_state['entities'],
+        mentions_extractor=mentions_extractor,
+        linker=linker
+    )
+    nlp.add_pipe("zshot", config=config, last=True)
+
+    return nlp
+
+predict = st.button("Run ZShot")
+if predict:
+    # placeholder for warning
+    placeholder = st.empty()
+    placeholder.info("Processing...")
+
+    nlp = build_pipeline()
+    doc = nlp(text)
+    placeholder.empty()
+
+    ent_html = displacy.render(doc, style="ent", jupyter=False)  # Display the entity visualization in the browser:
+    st.markdown(ent_html, unsafe_allow_html=True)
+
+st.sidebar.info("""See more: 
+- Check ZShot Github [here](https://github.com/IBM/zshot)
+- Check ZShot documentation [here](https://ibm.github.io/zshot/)""")

requirements.txt

@@ -0,0 +1,3 @@
+spacy
+zshot
+flair

requirements.txt.py

@@ -0,0 +1 @@
+zshot

streamlit_apps_config.py

@@ -0,0 +1,63 @@
+#APP STYLE
+MAX_WIDTH = 1600
+PADDING_TOP = 0.25
+PADDING_BOTTOM = 0.25
+PADDING_RIGHT = 4
+PADDING_LEFT = 4
+COLOR = 'black'
+BACKGROUND_COLOR = 'white'
+
+HTML_WRAPPER = """<div class="scroll entities" style="overflow-x: auto; border: 1px solid #e6e9ef; border-radius: 0.25rem; padding: 0.25rem; margin-bottom: 2.5rem; white-space:pre-wrap">{}</div>"""
+HTML_INDEX_WRAPPER = """<div ">{}</div>"""
+
+STYLE_CONFIG_OLD = f"""
+<style>
+    @import url('https://fonts.googleapis.com/css2?family=Montserrat:wght@300;400;500;600;700&display=swap');
+    *:not(text){{
+      font-family: Montserrat;
+    }}
+
+    .reportview-container .main .block-container{{
+        max-width: {MAX_WIDTH}px;
+        padding-top: {PADDING_TOP}rem;
+        padding-right: {PADDING_RIGHT}rem;
+        padding-left: {PADDING_LEFT}rem;
+        padding-bottom: {PADDING_BOTTOM}rem;
+    }}
+    .reportview-container .main {{
+        color: {COLOR};
+        background-color: {BACKGROUND_COLOR};
+    }}
+    div.scroll {{
+                margin:1px, 1px;
+                padding:1px;
+                width: 100%;
+                height: 500px;
+                overflow-x: hidden;
+                overflow-x: auto;
+    }}
+    .reportview-container .markdown-text-container{{
+        font-family: roboto !important;
+        color: dimgray !important;
+        line-height: normal !important;
+    }}
+    .reportview-container h2
+    {{
+        font-weight: 400 !important;
+        font-size: 1.5rem !important;
+        line-height: 1.6!important;
+    }}
+    .reportview-container h2
+    {{
+        font-weight: 300 !important;
+        font-size: 1.3rem !important;
+        line-height: 1.4!important;
+    }}
+
+
+</style>
+"""
+
+with open('./style.css') as f:
+    STYLE_CONFIG_NEW = f.read()
+STYLE_CONFIG = STYLE_CONFIG_OLD + '<style>{}</style>'.format(STYLE_CONFIG_NEW)

style.css

@@ -0,0 +1,292 @@
+@import url('https://fonts.googleapis.com/css2?family=Montserrat:wght@300;400;500;600;700&display=swap');
+
+header div[data-testid="stDecoration"] {
+    display: none;
+}
+
+/* sidebar */
+section[data-testid="stSidebar"] > div:first-child {
+    background: #ecf9ff;
+}
+
+section[data-testid="stSidebar"] div[data-testid="stMarkdownContainer"] h1 {
+	font-family: 'Montserrat', sans-serif;
+    font-weight: 500;
+    font-size: 18px;
+    line-height: 22px;
+    background: none;
+    color: #1E77B7;
+    margin-bottom: 10px;
+}
+
+section[data-testid="stSidebar"] button[title="View fullscreen"] {
+	right: 9px;
+	top: 5px;
+}
+
+section[data-testid="stSidebar"] button[kind="icon"] svg path,
+section[data-testid="stSidebar"] button[title="View fullscreen"] svg path {
+    fill: #3e4095;
+}
+
+section[data-testid="stSidebar"] button[title="View fullscreen"] svg {
+    position: relative;
+    top: -2px;
+}
+
+section[data-testid="stSidebar"] button[kind="icon"]:active {
+    background-color: #fff;
+}
+
+section[data-testid="stSidebar"] button[kind="icon"]:active,
+section[data-testid="stSidebar"] button[kind="icon"]:focus {
+    box-shadow: #3e4095 0px 0px 0px 0.2rem;
+    border-color: #3e4095;
+    color: #3e4095;
+}
+section[data-testid="stSidebar"] button[kind="icon"]:hover {
+    border-color: #3e4095;
+    color: #3e4095;
+}
+
+section[data-testid="stSidebar"] .stMultiSelect input[role="combobox"] {
+    display: none;
+}
+
+/* markdown container */
+div[data-testid="stMarkdownContainer"] {
+    font-family: Montserrat, sans-serif;
+}
+
+section[data-testid="stSidebar"] .element-container .stSelectbox label:empty {
+    display: none;
+}
+
+section[data-testid="stSidebar"] .element-container label {
+    margin-top: 50px;
+}
+
+section[data-testid="stSidebar"] div[data-testid="stTable"] {
+    padding: 0;
+}
+
+section[data-testid="stSidebar"] div[data-testid="stMarkdownContainer"] p {
+    margin: 0;
+}
+
+div[data-testid="stMarkdownContainer"] h1 {
+    font-family: Montserrat, sans-serif;
+    font-weight: 800;
+    font-size: 45px;
+    line-height: 55px;
+    color: #1E77B7;
+}
+
+div[data-testid="stMarkdownContainer"] h2 {
+    font-family: Montserrat, sans-serif;
+    font-weight: 500;
+    font-size: 18px;
+    line-height: 26px;
+    color: #1E77B7;
+}
+
+div[data-testid="stMarkdownContainer"] h3 {
+    font-family: Montserrat, sans-serif;
+    font-weight: 500;
+    font-size: 18px;
+    line-height: 22px;
+    color: #1E77B7;
+}
+
+div[data-testid="stMarkdownContainer"] code {
+    font-family: Montserrat, sans-serif;
+}
+
+/* selectbox */
+div[data-baseweb="select"] > div {
+    box-shadow: 0px 4px 10px rgba(0, 152, 218, 0.3);
+    border-radius: 2px;
+    min-height: 40px;
+    background: #ffffff;
+    border: none;
+}
+
+div[data-baseweb="select"] .st-bb.st-bc.st-b4.st-cy.st-by.st-cz.st-d0 svg {
+    fill: #BAD0DA;
+    height: 20px;
+    width: 20px;
+    padding: 0;
+}
+
+.stSelectbox > label {
+    font-family: 'Montserrat', sans-serif;
+    font-weight: 500;
+    font-size: 18px;
+    line-height: 22px;
+    color: #1E77B7;
+}
+
+div[data-baseweb="select"] > div > div > div {
+    font-family: Montserrat, sans-serif;
+    font-size: 14px !important;
+    line-height: 26px !important;
+    color: #536B76 !important;
+}
+
+div[role="listbox"] ul li {
+    color: rgb(128, 132, 149);
+}
+
+div[role="listbox"] ul li[aria-selected="true"],
+div[role="listbox"] ul li:hover {
+    color: #000;
+}
+
+/* multiselect */
+.stMultiSelect label {
+    font-family: 'Montserrat', sans-serif !important;
+    font-weight: 500;
+    font-size: 18px !important;
+    line-height: 22px;
+    background: none;
+    color: #1E77B7 !important;
+    margin-bottom: 10px !important;
+}
+
+.stMarkdown>div {
+    margin: 0;
+}
+
+.stMultiSelect span[data-baseweb="tag"] {
+    font-family: Montserrat, sans-serif;
+    background: #1E77B7;
+    color: #fff !important;
+    font-size: 12px !important;
+    line-height: 20px !important;
+}
+
+.stMultiSelect span[data-baseweb="tag"] span[role="presentation"] {
+    width: 21px;
+    height: 21px;
+    fill: #BAD0DA;
+}
+
+/* checkbox */
+label[data-baseweb="checkbox"] > span[role="checkbox"] {
+    border-color: #5334AE;
+    background-color: #5334AE;
+}
+
+label[data-baseweb="checkbox"] > div {
+}
+
+label[data-baseweb="checkbox"] label.st-ex {
+    background: red !important;
+}
+
+/* table */
+div[data-testid="stTable"] table {
+    width: 100%;
+    background: #ffffff;
+}
+
+div[data-testid="stTable"] th {
+    background: #3E4095;
+    vertical-align: middle;
+    color: #fff;
+    font-family: 'Montserrat', sans-serif;
+    font-weight: 800;
+    font-size: 16px;
+    line-height: 28px;
+    height: 30px;
+    padding: 10px;
+    text-align: left;
+    min-width: 103px;
+}
+
+div[data-testid="stTable"] tbody tr th,
+div[data-testid="stTable"] thead tr th:first-child {
+    display: none;
+}
+
+div[data-testid="stTable"] table tbody tr td {
+    font-family: Montserrat, sans-serif;
+    height: 50px;
+    padding: 10px;
+    font-size: 14px;
+    line-height: 20px;
+    color:#536B76;
+    border-bottom: 1px solid #E7EDF0;
+    text-align: left;
+}
+
+div[data-testid="stTable"] tr:nth-child(odd) {
+    box-shadow: inset 0px 0px 35px rgba(0, 152, 218, 0.05);
+}
+
+div[data-testid="stTable"] tr:nth-child(even) {
+    box-shadow: 0px 0px 35px rgba(0, 152, 218, 0.05);
+}
+
+/* entities */
+.scroll.entities {
+    border: 1px solid #E7EDF0;
+    border-radius: 3px;
+    text-align: justify;
+}
+
+.scroll.entities span {    
+    font-size: 14px;
+    line-height: 24px;
+    color: #536B76;
+    font-family: Montserrat, sans-serif;
+}
+.entity-wrapper {
+    border-radius: 3px;
+    padding: 1px;
+    margin: 0 2px 5px 2px;
+}
+
+.scroll.entities span .entity-type {
+    font-weight: 500;
+    color: #ffffff;
+    display: block;
+    padding: 3px 5px;
+}
+
+.scroll.entities span .entity-name {
+	border-radius: 3px;
+	padding: 2px 5px;
+	display: block;
+	margin: 3px 2px;
+}
+
+@media (max-width: 1199px) {
+    .reportview-container .main .block-container {
+        padding: 1rem;
+    }
+}
+
+@media (max-width: 991px) {
+    div[data-testid="stMarkdownContainer"] h1 {
+        font-size: 32px;
+        line-height: 38px;
+    }
+    section[data-testid="stSidebar"] > div:first-child {
+        padding: 2rem 1rem;
+    }
+    div[data-testid="stMarkdownContainer"] h2 {
+        padding-top: 0;
+        margin-top: 0;
+    }
+}
+
+@media (max-width: 767px) {
+    div[data-testid="stMarkdownContainer"] h1 {
+        font-size: 27px;
+        line-height: 33px;
+    }
+    .block-container .element-container .scroll.entities {
+        padding: 10px !important;
+    }
+}