update

app.py

@@ -1,56 +1,16 @@
-import streamlit as st
-import requests
-import plotly.express as px
-import pandas as pd
+from st_pages import Page, show_pages, add_page_title
 
-st.set_page_config(page_title="NicheImage Studio", layout="wide")
-st.markdown("## :black[Image Generation Studio by NicheImage]")
-replicate_logo = "assets/NicheTensorTransparent.png"
+# Optional -- adds the title and icon to the current page
+add_page_title()
 
-with st.sidebar:
-    st.image(replicate_logo, use_column_width=True)
-st.markdown(
-    """
-    **NicheImage is a decentralized network of image generation models, powered by the Bittensor protocol. Below you find information about the current models on the network.**
-    """,
-    unsafe_allow_html=True,
+# Specify what pages should be shown in the sidebar, and what their titles and icons
+# should be
+show_pages(
+    [
+        Page("home.py", "Home & Statistics", "📊"),
+        Page("pages/0_GoJourney.py", "GoJourney", "⛵"),
+        Page("pages/1_Text_To_Image.py", "Text To Image", "🎨"),
+        Page("pages/2_Image_To_Image.py", "Image To Image", "🖼️"),
+        Page("pages/3_Control_To_Image.py", "Control To Image", "🎛️"),
+    ]
 )
-response = requests.get(
-    "http://proxy_client_nicheimage.nichetensor.com:10003/get_uid_info"
-)
-if response.status_code == 200:
-    response = response.json()
-    # Plot distribution of models
-    model_distribution = {}
-    for uid, info in response["all_uid_info"].items():
-        model_name = info["model_name"]
-        model_distribution[model_name] = model_distribution.get(model_name, 0) + 1
-    fig = px.pie(
-        values=list(model_distribution.values()),
-        names=list(model_distribution.keys()),
-        title="Model Distribution",
-    )
-    st.plotly_chart(fig)
-    transformed_dict = []
-    for k, v in response["all_uid_info"].items():
-        transformed_dict.append(
-            {
-                "uid": k,
-                "model_name": v["model_name"],
-                "mean_score": (
-                    sum(v["scores"]) / (len(v["scores"])) if len(v["scores"]) > 0 else 0
-                ),
-            }
-        )
-    transformed_dict = pd.DataFrame(transformed_dict)
-    # plot N bar chart for N models, sorted by mean score
-    for model in model_distribution.keys():
-        model_data = transformed_dict[transformed_dict["model_name"] == model]
-        model_data = model_data.sort_values(by="mean_score", ascending=False)
-        if model_data.mean_score.sum() == 0:
-            continue
-        st.write(f"Model: {model}")
-        st.bar_chart(model_data[["uid", "mean_score"]].set_index("uid"))
-
-else:
-    st.error("Error getting miner info")

home.py

@@ -0,0 +1,56 @@
+import streamlit as st
+import requests
+import plotly.express as px
+import pandas as pd
+
+st.set_page_config(page_title="NicheImage Studio", layout="wide")
+st.markdown("## :black[Image Generation Studio by NicheImage]")
+replicate_logo = "assets/NicheTensorTransparent.png"
+
+with st.sidebar:
+    st.image(replicate_logo, use_column_width=True)
+st.markdown(
+    """
+    **NicheImage is a decentralized network of image generation models, powered by the Bittensor protocol. Below you find information about the current models on the network.**
+    """,
+    unsafe_allow_html=True,
+)
+response = requests.get(
+    "http://proxy_client_nicheimage.nichetensor.com:10003/get_uid_info"
+)
+if response.status_code == 200:
+    response = response.json()
+    # Plot distribution of models
+    model_distribution = {}
+    for uid, info in response["all_uid_info"].items():
+        model_name = info["model_name"]
+        model_distribution[model_name] = model_distribution.get(model_name, 0) + 1
+    fig = px.pie(
+        values=list(model_distribution.values()),
+        names=list(model_distribution.keys()),
+        title="Model Distribution",
+    )
+    st.plotly_chart(fig)
+    transformed_dict = []
+    for k, v in response["all_uid_info"].items():
+        transformed_dict.append(
+            {
+                "uid": k,
+                "model_name": v["model_name"],
+                "mean_score": (
+                    sum(v["scores"]) / (len(v["scores"])) if len(v["scores"]) > 0 else 0
+                ),
+            }
+        )
+    transformed_dict = pd.DataFrame(transformed_dict)
+    # plot N bar chart for N models, sorted by mean score
+    for model in model_distribution.keys():
+        model_data = transformed_dict[transformed_dict["model_name"] == model]
+        model_data = model_data.sort_values(by="mean_score", ascending=False)
+        if model_data.mean_score.sum() == 0:
+            continue
+        st.write(f"Model: {model}")
+        st.bar_chart(model_data[["uid", "mean_score"]].set_index("uid"))
+
+else:
+    st.error("Error getting miner info")

requirements.txt

@@ -5,4 +5,5 @@ streamlit-image-select
 plotly
 pandas
 httpx
-aiohttp
+aiohttp
+st-pages

utils/icon.py

@@ -0,0 +1,15 @@
+import streamlit as st
+
+
+@st.cache_data
+def show_icon(emoji: str):
+    """Shows an emoji as a Notion-style page icon.
+
+    Args:
+        emoji (str): name of the emoji, i.e. ":balloon:"
+    """
+
+    st.write(
+        f'<span style="font-size: 78px; line-height: 1">{emoji}</span>',
+        unsafe_allow_html=True,
+    )