Update app.py

app.py

@@ -1,12 +1,9 @@
 import gradio as gr
 import pandas as pd
-import io
-import os
+import matplotlib.pyplot as plt
 from pathlib import Path
 from huggingface_hub import HfApi, Repository
-import matplotlib.pyplot as plt
 
-# Set a clean, sans-serif default font
 plt.rcParams.update({
     "font.family": "sans-serif",
     "font.size": 10,
@@ -20,26 +17,26 @@ def upload_csv(file):
     df = pd.read_csv(file.name)
     if "text" not in df.columns or "label" not in df.columns:
         return (
-            gr.update(visible=False),  # hide table
-            "❌ CSV must contain `text` and `label` columns.",
-            gr.update(visible=False),  # hide Save
-            gr.update(visible=False),  # hide Download CSV
-            gr.update(visible=False),  # hide Visualize
-            gr.update(visible=False),  # hide Push Accordion
+            None,  # table
+            "❌ CSV must contain 'text' and 'label' columns.",
+            gr.update(visible=False),  # save
+            gr.update(visible=False),  # download CSV
+            gr.update(visible=False),  # visualize
+            gr.update(visible=False),  # push accordion
         )
     df["label"] = df["label"].fillna("")
     return (
-        gr.update(value=df[["text","label"]], visible=True),
-        "✅ File uploaded — you can now annotate.",
+        df[["text","label"]],
+        "✅ File uploaded — you can now annotate and use the buttons below.",
         gr.update(visible=True),
         gr.update(visible=True),
         gr.update(visible=True),
         gr.update(visible=True),
     )
 
-def save_changes(edited):
+def save_changes(table):
     global df
-    df = pd.DataFrame(edited, columns=["text","label"])
+    df = pd.DataFrame(table, columns=["text","label"])
     return "💾 Changes saved."
 
 def download_csv():
@@ -48,10 +45,10 @@ def download_csv():
     df.to_csv(path, index=False)
     return path
 
-def create_distribution_figure(df_input):
-    counts = df_input["label"].value_counts().sort_values(ascending=False)
-    labels = counts.index.tolist()
-    values = counts.values.tolist()
+def create_distribution_figure():
+    global df
+    counts = df["label"].value_counts().sort_values(ascending=False)
+    labels, values = counts.index.tolist(), counts.values.tolist()
     
     fig, (ax_table, ax_bar) = plt.subplots(
         ncols=2,
@@ -59,31 +56,18 @@ def create_distribution_figure(df_input):
         figsize=(8, max(2, len(labels)*0.4)),
         tight_layout=True
     )
-    
     # Table
     ax_table.axis("off")
-    table_data = [[lab, cnt] for lab, cnt in zip(labels, values)]
-    tbl = ax_table.table(
-        cellText=table_data,
-        colLabels=["Label", "Count"],
-        cellLoc="center",
-        loc="center"
-    )
-    tbl.auto_set_font_size(False)
-    tbl.set_fontsize(10)
-    tbl.scale(1, 1.2)
-    
+    data = [[l,v] for l,v in zip(labels, values)]
+    tbl = ax_table.table(cellText=data, colLabels=["Label","Count"], loc="center")
+    tbl.auto_set_font_size(False); tbl.set_fontsize(10); tbl.scale(1,1.2)
     # Bar chart
-    ax_bar.barh(labels, values, color="#222222")
-    ax_bar.invert_yaxis()
-    ax_bar.set_xlabel("Count")
-    ax_bar.set_ylabel("")
-    
+    ax_bar.barh(labels, values, color="#222")
+    ax_bar.invert_yaxis(); ax_bar.set_xlabel("Count")
     return fig
 
 def visualize_and_download_chart():
-    global df
-    fig = create_distribution_figure(df)
+    fig = create_distribution_figure()
     out_path = "label_distribution.png"
     fig.savefig(out_path, dpi=150, bbox_inches="tight")
     return fig, out_path
@@ -94,12 +78,10 @@ def push_to_hub(repo_name, hf_token):
         api = HfApi()
         api.create_repo(repo_id=repo_name, token=hf_token,
                         repo_type="dataset", exist_ok=True)
-
         local_dir = Path(f"./{repo_name.replace('/','_')}")
         if local_dir.exists():
             for f in local_dir.iterdir(): f.unlink()
             local_dir.rmdir()
-
         repo = Repository(
             local_dir=str(local_dir),
             clone_from=repo_name,
@@ -113,46 +95,47 @@ def push_to_hub(repo_name, hf_token):
         return f"❌ Push failed: {e}"
 
 with gr.Blocks(theme=gr.themes.Default()) as app:
-    gr.Markdown("## 🏷️ Label It! Text Annotation Tool")
-    gr.Markdown("**Step 1:** Upload a `.csv` with **text** + **label** columns.")
+    gr.Markdown("## 🏷️ Label It! Text Annotation Tool\n"
+                "Upload a `.csv` (with **text** + **label** columns), "
+                "then annotate, export, visualize, or publish.")
     
-    # Step 1: Upload only
+    # Step 1: Upload
     with gr.Row():
-        upload_file = gr.File(label="📁 Upload CSV", file_types=[".csv"])
-        upload_btn  = gr.Button("Upload")
-
-    # Hidden until upload
-    df_table      = gr.Dataframe(headers=["text","label"], visible=False, interactive=True)
-    status        = gr.Textbox(label="Status")
-    save_btn      = gr.Button("💾 Save", visible=False)
-    download_btn  = gr.Button("⬇️ Download CSV", visible=False)
-    download_csv_out = gr.File(label="📥 Download CSV", visible=False)
-
-    visualize_btn = gr.Button("📊 Visualize Distribution", visible=False)
-    chart_plot    = gr.Plot(label="Label Distribution", visible=False)
-    download_chart= gr.File(label="📥 Download Chart", visible=False)
-
-    # Push accordion hidden
-    with gr.Accordion("📦 Push to Hugging Face Hub", open=False, visible=False) as push_acc:
-        repo_in  = gr.Textbox(label="Repo (username/dataset-name)")
-        token_in = gr.Textbox(label="🔑 HF Token", type="password")
-        push_btn = gr.Button("🚀 Push")
-        push_status = gr.Textbox(label="Push Status")
-
-    # Bind actions
+        file_input = gr.File(label="📁 Upload CSV", file_types=[".csv"])
+        upload_btn = gr.Button("Upload")
+    
+    # Editable table
+    table = gr.Dataframe(headers=["text","label"], interactive=True, visible=False)
+    status = gr.Textbox(label="Status", interactive=False)
+    
+    # Step 2 buttons (hidden initially)
+    with gr.Row(visible=False) as action_row:
+        save_btn     = gr.Button("💾 Save")
+        download_btn = gr.Button("⬇️ Download CSV")
+        visualize_btn= gr.Button("📊 Visualize Distribution")
+        download_csv_out   = gr.File(label="📥 Download CSV")
+        chart_plot         = gr.Plot(label="Label Distribution")
+        download_chart_out = gr.File(label="📥 Download Chart")
+    
+    # Push accordion
+    push_acc = gr.Accordion("📦 Push to Hugging Face Hub", open=False, visible=False)
+    with push_acc:
+        repo_in   = gr.Textbox(label="Repo (username/dataset-name)")
+        token_in  = gr.Textbox(label="🔑 HF Token", type="password")
+        push_btn  = gr.Button("🚀 Push")
+        push_status = gr.Textbox(label="Push Status", interactive=False)
+    
+    # Event bindings
     upload_btn.click(
         upload_csv,
-        inputs=upload_file,
-        outputs=[df_table, status,
+        inputs=file_input,
+        outputs=[table, status,
                  save_btn, download_btn, visualize_btn, push_acc]
     )
-    save_btn.click(save_changes, inputs=df_table, outputs=status)
+    save_btn.click(save_changes, inputs=table, outputs=status)
     download_btn.click(download_csv, outputs=download_csv_out)
     visualize_btn.click(visualize_and_download_chart,
-                        outputs=[chart_plot, download_chart])
+                        outputs=[chart_plot, download_chart_out])
     push_btn.click(push_to_hub, inputs=[repo_in, token_in], outputs=push_status)
 
-    # Step 2 instruction
-    gr.Markdown("**Step 2:** Edit labels, then Save, Visualize or Publish.")
-
 app.launch()