update app

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.html filter=lfs diff=lfs merge=lfs -text

app.py

@@ -1,13 +1,13 @@
-import pandas as pd
 import streamlit as st
 from st_on_hover_tabs import on_hover_tabs
 
 from st_pages.home import home_page
 from st_pages.analyse import analyse_page
+from st_pages.dataset import dataset_page
 from st_pages.dashboard import dashboard
 from st_pages.about import about_us_page
-from utils.constants import PAGE_FAVICON, CONTRIBUTORS
-from utils.utils import load_css, init_session_state
+from utils.constants import PAGE_FAVICON, CONTRIBUTORS, DEFAULT_CHOICE
+from utils.utils import load_css, init_session_state, load_header
 
 st.set_page_config(
     page_title="Sentiment Analysis Tool", page_icon=PAGE_FAVICON, layout="wide"
@@ -19,33 +19,36 @@ init_session_state()
 with st.sidebar:
     st.write("<br>" * 4, unsafe_allow_html=True)
     selected_task = on_hover_tabs(
-        tabName=["Home Page", "Analyse Sentiment", "Dashboard", "About Us"],
-        iconName=["home", "engineering", "insert_chart", "contact_support"],
+        tabName=["Home Page", "Analyse Sentiment", "Dashboard", "Data", "About Us"],
+        iconName=["home", "engineering", "equalizer", "analytics", "contact_support"],
         styles={
             "navtab": {"background-color": "#fff"},
             "tabOptionsStyle": {
                 ":hover :hover": {"color": "#170034", "cursor": "pointer"}
             },
         },
-        default_choice=0,
+        default_choice=DEFAULT_CHOICE,
     )
 
 if selected_task == "Home Page":
-    home_page()
+    with st.spinner("Loading Home Page..."):
+        home_page()
 
 elif selected_task == "Analyse Sentiment":
-    analyse_page()
+    with st.spinner("Loading Analyse Page..."):
+        analyse_page()
 
 elif selected_task == "Dashboard":
     if "master_df" in st.session_state and st.session_state["master_df"] is None:
-        # load_header("Sentiment Analysis Dashboard")
-        # st.info("Please analyze a tweet first.")
-        st.session_state["master_df"] = pd.read_csv(
-            "assets/dataset/temp_output_combined.csv"
-        )
-    # else:
-    with st.spinner("Loading Dashboard..."):
-        dashboard()
+        load_header("Sentiment Analysis Dashboard")
+        st.info("Please analyze a tweet first.")
+    else:
+        with st.spinner("Loading Dashboard..."):
+            dashboard()
+
+elif selected_task == "Data":
+    load_header("Manually Labelled Dataset")
+    dataset_page()
 
 elif selected_task == "About Us":
     about_us_page(CONTRIBUTORS)

assets/html/about.html

@@ -1,9 +1,3 @@
-<div style="text-align: center">
-  <br>
-  <table>
-    <tbody class="table">
-      <th width="20%" colspan="3">Team Members</th>
-      <tr width="20%">{team}</tr>
-    </tbody>
-  </table>
-</div>
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6fdd84a069b0f6cd27c32dc190a8dcc9a60644edd9eb87db92193619209c892
+size 200

assets/html/report.html

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cb4288084f42117df155d8cc4ff449c341fd3e2bd81bdbad8707a217cfba88b
+size 12001413

st_pages/about.py

@@ -4,7 +4,7 @@ from utils.constants import CONTRIBUTORS
 
 
 def about_us_page(contributors):
-    load_header("Meet our CONTRIBUTORS")
+    load_header("Meet Our Team")
 
     data = contributors
     full_string = ""

st_pages/analyse.py

@@ -1,3 +1,4 @@
+import pandas as pd
 import streamlit as st
 from utils.scraper import fetch_main_tweet_dataframe, fetch_comments_dataframe
 from utils.utils import (
@@ -33,20 +34,28 @@ def analyse_page():
             df_author = fetch_main_tweet_dataframe(twitter_url)
             df_comments = fetch_comments_dataframe(twitter_url)
 
-            # df_author = pd.read_csv('assets/dataset/temp_output_author.csv')
-            # df_comments = pd.read_csv('assets/dataset/temp_output_comments.csv')
+            df_author = pd.read_csv("assets/dataset/temp_output_author.csv")
+            df_comments = pd.read_csv("assets/dataset/temp_output_comments.csv")
 
             master_df = combine_author_and_comments_df(df_author, df_comments)
             st.session_state["master_df"] = master_df
 
-            # master_df = pd.read_csv('assets/dataset/temp_output_combined2.csv')
-            # master_df = master_df.astype({'id': str, 'inReplyToId': str})
-            # st.session_state['master_df'] = master_df
+    elif not submitted:
+        with st.spinner("Loading data..."):
+            df_author = pd.read_csv("assets/dataset/temp_output_author.csv")
+            df_comments = pd.read_csv("assets/dataset/temp_output_comments.csv")
 
-    if "master_df" in st.session_state and st.session_state["master_df"] is not None:
-        st.dataframe(
-            st.session_state["master_df"], height=450, use_container_width=True
-        )
+            master_df = combine_author_and_comments_df(df_author, df_comments)
+            st.session_state["master_df"] = master_df
+
+        with st.expander("Sample Data", expanded=True):
+            if (
+                "master_df" in st.session_state
+                and st.session_state["master_df"] is not None
+            ):
+                st.dataframe(
+                    st.session_state["master_df"], height=450, use_container_width=True
+                )
 
         with cols[2]:
             st.write("<br>", unsafe_allow_html=True)

st_pages/dashboard.py

@@ -1,11 +1,19 @@
-import pandas as pd
 import streamlit as st
 from utils.generate_pdf import construct_pdf
 from utils.utils import load_header, add_columns_for_graphs
-from utils.graph_functions import *
+from utils.graph_functions import (
+    metrics_bar,
+    most_common_trigrams,
+    display_word_cloud,
+    crear_grafico_dispersion,
+    display_target_count,
+    sentiment_over_date,
+    stacked_bar_fig,
+)
 
 
 def dashboard():
+    # st.session_state["master_df"] = pd.read_csv("output.csv")
     master_df = st.session_state["master_df"]
 
     if st.session_state["tweet_data"] is None:
@@ -101,13 +109,13 @@ if __name__ == "__main__":
     )
     init_session_state()
 
-    df = pd.read_csv("assets/dataset/master.csv")
-    st.session_state["master_df"] = pd.read_csv(
-        "assets/dataset/temp_output_combined.csv"
-    )
+    # df = pd.read_csv("assets/dataset/master.csv")
+    # st.session_state["master_df"] = pd.read_csv(
+    #     "assets/dataset/temp_output_combined.csv"
+    # )
 
-    df_author = pd.read_csv("assets/dataset/temp_output_author.csv")
-    df_comments = pd.read_csv("assets/dataset/temp_output_comments.csv")
+    # df_author = pd.read_csv("assets/dataset/temp_output_author.csv")
+    # df_comments = pd.read_csv("assets/dataset/temp_output_comments.csv")
 
     # with st.spinner("Analyzing Tweets..."):
     #     master_df, tweet_data = construct_master_df(df_author, df_comments)

st_pages/dataset.py

@@ -0,0 +1,44 @@
+import pandas as pd
+import streamlit as st
+import streamlit.components.v1 as components
+
+
+def dataset_page():
+    df = pd.read_csv("assets/dataset/manually_labelled/output_raw.csv", index_col=0)
+
+    df["comment_id"] = df["comment_id"].astype("str")
+    df["post_id"] = df["post_id"].astype("str")
+
+    left, _, right = st.columns([2, 2, 1])
+
+    with left:
+        st.caption(
+            "Our team developed manually labelled datasets for the purpose of training our model. The dataset contains 3000 tweets. The dataset is divided into three subsets: train, test, and validation."
+        )
+
+    with right:
+        st.write("<br>", unsafe_allow_html=True)
+        with open(
+            "assets/dataset/manually_labelled/omdena-IREX-manually-labelled-dataset.zip",
+            "rb",
+        ) as fp:
+            st.download_button(
+                label="Download Datasets",
+                data=fp,
+                file_name="omdena-IREX-manually-labelled-dataset.zip",
+                mime="application/zip",
+                use_container_width=True,
+            )
+
+    with st.expander("Dataset Description"):
+        components.html(
+            open("assets/html/report.html", "r", encoding="utf8").read(), height=17450
+        )
+
+    with st.expander("Manually Labelled Dataset (Merged)", expanded=True):
+        st.dataframe(df, height=300)
+
+
+if __name__ == "__main__":
+    st.set_page_config(page_title="Sentiment Analysis", layout="wide")
+    dataset_page()

utils/constants.py

@@ -1,3 +1,7 @@
+from sklearn.feature_extraction import text as sklearn_text
+
+DEFAULT_CHOICE = 0
+
 PAGE_BANNER = "assets/img/banner.png"
 PAGE_FAVICON = "assets/img/favicon.png"
 PAGE_BACKGROUND = "assets/img/background.jpg"
@@ -32,9 +36,6 @@ CONTRIBUTORS = [
 
 TEXT_COLOR = {"positive": "green", "neutral": "gray", "negative": "red"}
 
-
-from sklearn.feature_extraction import text as sklearn_text
-
 spanish_stop_words = sklearn_text.ENGLISH_STOP_WORDS.union(
     set(
         [