Update app.py

app.py

@@ -5,7 +5,7 @@ import pandas as pd
 from pathlib import Path
 import matplotlib
 import numpy as np
-import gradio
+import gradio as gr
 
 ############################################################# 2D Line Plot ########################################################
 ### dvq stuff, obvs this will just be an import in the final version
@@ -883,6 +883,39 @@ with ui.navset_card_tab(id="tab"):
             return fig
     with ui.nav_panel("Viral Model"):
         gr.load("models/Hack90/virus_pythia_31_1024").launch()
+
+    with ui.nav_panel("Viral Model Training"):    
+        ui.page_opts(fillable=True)
+        ui.panel_title("Does context size matter for a nucleotide model?")
+        
+        def plot_loss_rates(df, type): 
+            # interplot each column to be same number of points
+            x = np.linspace(0, 1, 1000)
+            loss_rates = []
+            labels = ['32', '64', '128', '256', '512', '1024']
+            #drop the column step
+            df = df.drop(columns=['Step'])
+            for col in df.columns:
+                y = df[col].dropna().astype('float', errors = 'ignore').dropna().values
+                f = interp1d(np.linspace(0, 1, len(y)), y)
+                loss_rates.append(f(x))
+            fig, ax = plt.subplots()
+            for i, loss_rate in enumerate(loss_rates):
+                ax.plot(x, loss_rate, label=labels[i])
+            ax.legend()
+            ax.set_title(f'Loss rates for a {type} parameter model')
+            ax.set_xlabel('Training steps')
+            ax.set_ylabel('Loss rate')
+            return fig
+            
+        @render.plot
+        def plot():
+            fig = None
+            df = pd.read_csv('14m.csv')
+            mpl.rcParams.update(mpl.rcParamsDefault)
+            fig = plot_loss_rates(df, '14M')
+            return fig
+        
 # @render.image  
 # def image():
 #     img = None