app.py

import gradio as gr
import pandas as pd
import numpy as np
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score, balanced_accuracy_score, precision_score, recall_score, roc_auc_score
from sklearn.calibration import calibration_curve
import matplotlib.pyplot as plt
import seaborn as sns
from io import StringIO
import warnings
warnings.filterwarnings('ignore')
import numpy as np
import pandas as pd
import pyarrow.parquet as pq
from sklearn.preprocessing import OneHotEncoder,MinMaxScaler
from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier
from sklearn.model_selection import train_test_split,cross_val_score,StratifiedKFold,RepeatedStratifiedKFold
from sklearn.metrics import confusion_matrix,classification_report,precision_score, recall_score, f1_score, accuracy_score, balanced_accuracy_score, matthews_corrcoef
from sklearn.metrics import roc_auc_score,auc
import pickle

from sklearn.utils.class_weight import compute_sample_weight

import xgboost as xgb
from xgboost.sklearn import XGBClassifier
from sklearn.naive_bayes import GaussianNB
from sklearn.ensemble import AdaBoostClassifier
from sklearn.svm import SVC
from sklearn.linear_model import LogisticRegression
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import brier_score_loss
from sklearn.calibration import calibration_curve
import matplotlib.pyplot as plt
from sklearn.calibration import CalibratedClassifierCV
from sklearn.linear_model import LinearRegression

# Global variables for training data and column names
training_data = None
column_names = None
test_list=[]
def rand_for(neww_list,x_te,rf,lab,x_tr,actual,paramss,X_Tempp,enco,my_table_str,my_table_num,tabl,tracount):
  cl_list=[]
  pro_list=[]
  for i in neww_list:
    dff_copy=i.copy()
    y_cl=dff_copy.loc[:,lab]
    teemp_list=[]
    ftli=[]
    X_cl=dff_copy.drop([lab],axis=1)
    x_te=pd.DataFrame(x_te,columns=X_Tempp.columns)

    if tracount==0:

      #mm=RandomForestClassifier(n_estimators=100, criterion='entropy',random_state=42,bootstrap=True, oob_score=True,class_weight='balanced',ccp_alpha=0.01)
      mm=RandomForestClassifier(n_estimators=100, criterion='entropy',max_features=None,random_state=42,bootstrap=True, oob_score=True,class_weight='balanced',ccp_alpha=0.01)
      #mm.fit(X_cl,y_cl)
      calibrated_rf = CalibratedClassifierCV(estimator=mm, method='isotonic', cv=5)
      calibrated_rf.fit(X_cl, y_cl)
      #print(calibrated_rf.get_params())
      out=calibrated_rf.predict(x_te)
      probs=calibrated_rf.predict_proba(x_te)[:,1]
    elif tracount==1:
      dtrain = xgb.DMatrix(X_cl.to_numpy(), label=y_cl)
      dtest = xgb.DMatrix(x_te.to_numpy(), label=y_te)
      params = {
        'objective': 'binary:logistic',  # Binary classification problem
        'eval_metric': 'logloss',  # Logarithmic loss for evaluation
        'max_depth': 60,
        'eta': 0.1,
        'subsample': 0.8,
        'colsample_bytree': 0.8,
        'seed': 42}
      num_rounds = 100
      mm=xgb.train(params, dtrain, num_rounds)
      probs = mm.predict(dtest)
      out = (probs > 0.5).astype(int)

    elif tracount==5:
      mm=LogisticRegression(penalty='l2',solver='newton-cholesky',max_iter=200)
      mm.fit(X_cl,y_cl)
      out=mm.predict(x_te)
      probs=mm.predict_proba(x_te)[:,1]

  
    elif tracount==4:
      var_smoothing_value = 1e-9  # Adjust this value as needed
      mm = GaussianNB(var_smoothing=var_smoothing_value)
      mm.fit(X_cl, y_cl)
      out = mm.predict(x_te)
      probs = mm.predict_proba(x_te)[:, 1]

    elif tracount==1:
      mm = AdaBoostClassifier(n_estimators=100,random_state=42,estimator=RandomForestClassifier(n_estimators=100, criterion='entropy',random_state=42,bootstrap=True, oob_score=True,class_weight='balanced',ccp_alpha=0.01))
      out = mm.predict(x_te)
      probs = mm.predict_proba(x_te)[:, 1]

    elif tracount==6:
      mm = SVC(probability=True, C=3)
      mm.fit(X_cl, y_cl)
      out = mm.predict(x_te)
      probs = mm.predict_proba(x_te)[:, 1]



    cl_list.append(out)
    pro_list.append(probs)
    
   
  
  return cl_list,pro_list
def ne_calib(some_prob,down_factor,origin_factor):
  aa=some_prob*origin_factor/down_factor
  denone=(1-some_prob)*(1-origin_factor)/(1-down_factor)
  new_dum_prob=aa/(denone+aa)
  return new_dum_prob
def actualll(sl_list,pro_list,delt,down_factor,origin_factor):
  ac_list=[]
  probab_list=[]
  second_probab_list=[]

  for i in range(len(sl_list[0])):
    sum=0
    sum_pro=0
    sum_pro_pro=0
    for j in range(len(sl_list)):

      sum_pro+=ne_calib(pro_list[j][i],down_factor,origin_factor)
      sum_pro_pro+=pro_list[j][i]

      if sl_list[j][i]==-1:
        sum+=(sl_list[j][i])
      else:
        sum+=(sl_list[j][i])
   
    sum/=len(sl_list)
    sum_pro/=len(sl_list)
    sum_pro_pro/=len(sl_list)


    if sum>=delt:
      ac_list.append(1)
      probab_list.append(sum_pro)
      second_probab_list.append(sum_pro_pro)
    elif sum<=delt and sum >=0 :
      ac_list.append(0)
      probab_list.append(1-sum_pro)
      second_probab_list.append(1-sum_pro_pro)
    elif sum<=delt and sum <0:
      ac_list.append(0)
      probab_list.append(sum_pro)
      second_probab_list.append(sum_pro_pro)
  return ac_list,probab_list,second_probab_list
 


def sli_mod(c_lisy):
  sli_list=[]
  ### I am changing the threshold
  for i in c_lisy:
    k=np.array(i)
    k[k<0.5]=-1
    k[k>=0.5]=1
    #k=k/len(c_lisy)
    sli_list.append(list(k))
  return sli_list

def run_model(x_tr,x_te,y_tr,deltaa,lab,rf,X_Tempp,track,actual,paramss,enco,my_table_str,my_table_num,tabl,tracount,origin_factor):

  x_tr=pd.DataFrame(x_tr,columns=X_Tempp.columns)
  y_tr=pd.DataFrame(y_tr,columns=[test_list[track]])
  master_table=pd.concat([x_tr,y_tr],axis=1).copy()

  only_minority=master_table.loc[master_table[lab]==1]

  only_majority=master_table.drop(only_minority.index)
  min_index=only_minority.index
  max_index=only_majority.index

  df_list=[]
  down_factor=0
  if (len(min_index)<=60):
    for i in range(20):
      np.random.seed(i+30)
      if test_list[track]=='VOD' or test_list[track]=='STROKEHI':# or test_list[track]=='ACSPSHI' or test_list[track]=='AVNPSHI':
        sampled_array = np.random.choice(max_index,size=int(3*len(min_index)), replace=True)
        down_factor=0.25
      elif test_list[track]=='ACSPSHI':
        sampled_array = np.random.choice(max_index,size=int(2.5*len(min_index)), replace=True)
        down_factor=1/(1+2.5)
      else:
        sampled_array = np.random.choice(max_index,size=int(2*len(min_index)), replace=True)
        down_factor=1/(1+2)
      temp_df=only_majority.loc[sampled_array]

      new_df=pd.concat([temp_df,only_minority])
 
      df_list.append(new_df)
  else:
    for i in range(10):
      np.random.seed(i+30)
      if test_list[track]=='DEAD':
        sampled_array = np.random.choice(max_index,size=int(3*len(min_index)), replace=True)
        down_factor=1/(1+3)
      else:
        sampled_array = np.random.choice(max_index,size=int(3*len(min_index)), replace=True)
        down_factor=1/(1+3)
      temp_df=only_majority.loc[sampled_array]
 
      new_df=pd.concat([temp_df,only_minority])
 
      df_list.append(new_df)


      
  #neww_list=my_tomek(df_list,lab)
  neww_list=df_list
  c_lisy,pro_lisy=rand_for(neww_list,x_te,rf,lab,x_tr,actual,paramss,X_Tempp,enco,my_table_str,my_table_num,tabl,tracount)
  sli_lisy=sli_mod(c_lisy)

  a_lisy,probab_lisy,secondlisy=actualll(sli_lisy,pro_lisy,deltaa,down_factor,origin_factor)
  return a_lisy,probab_lisy,secondlisy
def load_training_data():

    global training_data, column_names, test_list
    

    try:
      my_table=pq.read_table('year6.parquet').to_pandas()
      print(my_table['YEARGPF'].value_counts())
      my_table=my_table[(my_table['YEARGPF']!='< 2008')]
      my_table=my_table.reset_index(drop=True)
 
      pa=pd.read_csv('final_variable.csv')
      pali=list(pa.iloc[:,0])
      print(pali)

      #pali.append(test_list[track])
      #pali.append('DUMMYID')
      #pali.remove('AGEGPFF')
      #pali.remove('COUNTRY')
      #print(pali)
      #my_table=my_table[pali]
      training_data = my_table
      column_names=pali
    except FileNotFoundError:

        return "No training Data"

def train_and_evaluate(input_file):
    
    global training_data, column_names,test_list
    
    if training_data is None or column_names is None:
        load_training_data()
    
    if input_file is None:
        return None, None, None
    
    try:
  
        input_data = pd.read_csv(input_file.name)
        

        available_features = [col for col in column_names if col in training_data.columns]
        available_features_input = [col for col in available_features if col in input_data.columns]
        
        if not available_features_input:
            return "Error: No matching columns found between datasets", None, None
        
        # Prepare training data
        
        #X_train_full = training_data[available_features]
        outcome_cols = ['DEAD', 'GF', 'AGVHD', 'CGVHD', 'VOCPSHI', 'STROKEHI']
        test_list=outcome_cols.copy()
        total_cols=available_features+outcome_cols
        inter_df=training_data[total_cols]
        inter_df=inter_df.dropna()
        inter_df=inter_df.reset_index(drop=True)


        input_data=input_data[(input_data['YEARGPF']!='< 2008')]
        input_data=input_data.reset_index(drop=True)

        inter_input=input_data[total_cols]
        inter_input=inter_input.dropna()
        inter_input=inter_input.reset_index(drop=True)
        my_table=inter_df[available_features]
        # Prepare input data
        X_input = inter_input[available_features]
        X_input = X_input[my_table.columns]
        my_test=X_input
        '''li1=['Yes','No']
        li2=['Event happened', 'No event']
        cols_with_unique_values1 = []
        cols_with_unique_values2 = []
        #print(my_table['EXCHTFPR'].isin(li1))
        for col in my_table.columns:
          if my_table[col].isin(li1).all():
              cols_with_unique_values1.append(col)
        for col in my_table.columns:
          if my_table[col].isin(li2).all():
              cols_with_unique_values2.append(col)
        #print(len(cols_with_unique_values1))
        #print(len(cols_with_unique_values2))
        my_ye=my_table[cols_with_unique_values1].replace(['Yes','No'],[1,0]).astype('int64')
        my_eve=my_table[cols_with_unique_values2].replace(['Event happened','No event'],[1,0]).astype('int64')
        my_table2=my_table.copy()
        ccc=[elem for elem in  cols_with_unique_values1+cols_with_unique_values2]
        #print(ccc)
        my_table_modify=my_table2.drop(ccc,axis=1)
        my_table_modify=pd.concat([my_table_modify,my_ye,my_eve],axis=1)
        #my_table_modify=my_table_modify.drop([test_list[track],'DUMMYID'],axis=1)
        my_table_str=my_table_modify.select_dtypes(exclude=['number'])
        print(my_table_str.shape)
        my_table_num=my_table_modify.select_dtypes(include=['number'])
        #print(my_table_num.shape)
        enco=OneHotEncoder(sparse_output=True)
        fito=enco.fit(my_table_str)
        #mmm=aa.inverse_transform(g)
        tabl=enco.transform(my_table_str)
        tabl=pd.DataFrame(tabl.toarray(),columns=enco.get_feature_names_out())
        #print(tabl.shape)
        #print(dfcopy)
        ftable=pd.concat([tabl,my_table_num],axis=1)
        X_train_full=ftable
        li1=['Yes','No']
        li2=['Event happened', 'No event']
        cols_with_unique_values1 = []
        cols_with_unique_values2 = []
        for col in my_test.columns:
          if my_test[col].isin(li1).all():
              cols_with_unique_values1.append(col)
        for col in my_test.columns:
          if my_test[col].isin(li2).all():
              cols_with_unique_values2.append(col)
        #print(len(cols_with_unique_values1))
        #print(len(cols_with_unique_values2))
        my_ye=my_test[cols_with_unique_values1].replace(['Yes','No'],[1,0]).astype('int64')
        my_eve=my_test[cols_with_unique_values2].replace(['Event happened','No event'],[1,0]).astype('int64')
        my_test2=my_test.copy()
        ccc=[elem for elem in  cols_with_unique_values1+cols_with_unique_values2]
        #print(ccc)
        my_test_modify=my_test2.drop(ccc,axis=1)
        my_test=pd.concat([my_test_modify,my_ye,my_eve],axis=1)
        #print(my_table_str.shape)
        my_test_num=my_test.select_dtypes(include=['number'])
        my_test_str=my_test.select_dtypes(exclude=['number'])
        mm=my_test_str.columns
        my_test_str=enco.transform(my_test_str)
        my_test_str=pd.DataFrame(my_test_str.toarray(),columns=enco.get_feature_names_out())
        my_test_real=pd.concat([my_test_str,my_test_num],axis=1)'''

        # Train data numerical
        li1=['Yes','No']
        #li2=['Event happened', 'No event']
        cols_with_unique_values1 = []
        cols_with_unique_values2 = []
        #print(my_table['EXCHTFPR'].isin(li1))
        for col in my_table.columns:
          if my_table[col].isin(li1).all():
              cols_with_unique_values1.append(col)
        #for col in my_table.columns:
          #if my_table[col].isin(li2).all():
              #cols_with_unique_values2.append(col)
        #print(len(cols_with_unique_values1))
        #print(len(cols_with_unique_values2))
        my_ye=my_table[cols_with_unique_values1].replace(['Yes','No'],[1,0]).astype('int64')
        #my_eve=my_table[cols_with_unique_values2].replace(['Event happened','No event'],[1,0]).astype('int64')
        my_table2=my_table.copy()
        ccc=[elem for elem in  cols_with_unique_values1+cols_with_unique_values2]
        #print(ccc)
        my_table_modify=my_table2.drop(ccc,axis=1)
        my_table_modify=pd.concat([my_table_modify,my_ye],axis=1)
        #my_table_modify=my_table_modify.drop([test_list[track],'DUMMYID'],axis=1)
        my_table_str=my_table_modify.select_dtypes(exclude=['number'])
        print(my_table_str.shape)
        my_table_num=my_table_modify.select_dtypes(include=['number'])

        #Test Data Numerical
        li1=['Yes','No']
        li2=['Event happened', 'No event']
        cols_with_unique_values1 = []
        cols_with_unique_values2 = []
        for col in my_test.columns:
          if my_test[col].isin(li1).all():
              cols_with_unique_values1.append(col)
        for col in my_test.columns:
          if my_test[col].isin(li2).all():
              cols_with_unique_values2.append(col)
        #print(len(cols_with_unique_values1))
        #print(len(cols_with_unique_values2))
        my_ye=my_test[cols_with_unique_values1].replace(['Yes','No'],[1,0]).astype('int64')
        #my_eve=my_test[cols_with_unique_values2].replace(['Event happened','No event'],[1,0]).astype('int64')
        my_test2=my_test.copy()
        ccc=[elem for elem in  cols_with_unique_values1+cols_with_unique_values2]
        #print(ccc)
        my_test_modify=my_test2.drop(ccc,axis=1)
        my_test=pd.concat([my_test_modify,my_ye],axis=1)
        #print(my_table_str.shape)
        my_test_num=my_test.select_dtypes(include=['number'])
        my_test_str=my_test.select_dtypes(exclude=['number'])
        mm=my_test_str.columns


        # Common encoding
        df_combined = pd.concat([my_table_str, my_test_str], axis=0, ignore_index=True)
        enco = OneHotEncoder(sparse_output=False, handle_unknown='ignore') 
        encoded = enco.fit_transform(df_combined)
        encoded_df = pd.DataFrame(encoded, columns=enco.get_feature_names_out())

        tabl = encoded_df.iloc[:len(my_table_str)].reset_index(drop=True)
        tabl=tabl.reset_index(drop=True)
        ftable=pd.concat([tabl,my_table_num],axis=1)
        X_train_full=ftable
        my_test_str = encoded_df.iloc[len(my_table_str):].reset_index(drop=True)
        my_test_str=my_test_str.reset_index(drop=True)
        my_test_real=pd.concat([my_test_str,my_test_num],axis=1)

        
   

        metrics_results = []
        calibration_results = []
        calibration_plots = []
        
        outcome_names = ['Overall Survival', 'Graft Failure', 'Acute GVHD', 'Chronic GVHD', 'Vaso-Occlusive Crisis Post-HCT', 'Stroke Post-HCT']
        
        for i, (outcome_col, outcome_name) in enumerate(zip(outcome_cols, outcome_names)):
            if outcome_col not in training_data.columns:
                continue
                
            y_train_full = inter_df[outcome_col]
            amaj1=y_train_full.value_counts().idxmax()
            amin1=y_train_full.value_counts().idxmin()
            #print(y.value_counts().idxmax())
            y_train_full=y_train_full.replace([amin1,amaj1],[1,0]).astype(int)  # FIX 1: added .astype(int)

            y_test_full = inter_input[outcome_col]
            amaj1=y_test_full.value_counts().idxmax()
            amin1=y_test_full.value_counts().idxmin()
            #print(y.value_counts().idxmax())
            y_test_full=y_test_full.replace([amin1,amaj1],[1,0]).astype(int)  # FIX 1: added .astype(int)
            
            X_train,y_train=X_train_full.values,y_train_full.values
            x_te,y_test=my_test_real.values,y_test_full.values
            vddc=len(np.where(y_train_full.to_numpy()==1)[0])/X_train_full.shape[0]
            deltaa=0.2
            rf=RandomForestClassifier()
            paramss={}
            tracount=0
            y_pred,y_pred_proba,secondnaive=run_model(X_train,x_te,y_train,deltaa,outcome_col,rf,X_train_full,i,ftable,paramss,enco,my_table_str,my_table_num,tabl,tracount,vddc)
            #mm=RandomForestClassifier(n_estimators=100, criterion='entropy')
            #mm.fit(X_train,y_train)
            #y_pred=mm.predict(x_te)
            #y_pred_proba=mm.predict_proba(x_te)[:,1]

            accuracy = accuracy_score(y_test, y_pred)
            balanced_acc = balanced_accuracy_score(y_test, y_pred)
            precision = precision_score(y_test, y_pred, average='weighted', zero_division=0)
            recall = recall_score(y_test, y_pred, average='weighted', zero_division=0)
            auc = roc_auc_score(y_test, y_pred_proba)
            
            metrics_results.append([outcome_name, f"{accuracy:.3f}", f"{balanced_acc:.3f}", 
                                  f"{precision:.3f}", f"{recall:.3f}", f"{auc:.3f}"])
            

            fraction_pos, mean_pred = calibration_curve(y_test, y_pred_proba, n_bins=10)
            

            if len(mean_pred) > 1 and len(fraction_pos) > 1:
                slope = np.polyfit(mean_pred, fraction_pos, 1)[0]
                intercept = np.polyfit(mean_pred, fraction_pos, 1)[1]
            else:
                slope, intercept = 1.0, 0.0
            
            calibration_results.append([outcome_name, f"{slope:.3f}", f"{intercept:.3f}"])
            

            fig, ax = plt.subplots(figsize=(8, 6))
            ax.plot([0, 1], [0, 1], 'k--', label='Perfect Calibration')
            ax.plot(mean_pred, fraction_pos, 'o-', label=f'{outcome_name}')
            ax.set_xlabel('Mean Predicted Probability')
            ax.set_ylabel('Fraction of Positives')
            ax.set_title(f'Calibration Plot - {outcome_name}')
            ax.legend()
            ax.grid(True, alpha=0.3)
            plt.tight_layout()
            calibration_plots.append(fig)
        

        metrics_df = pd.DataFrame(metrics_results, 
                                columns=['Outcome', 'Accuracy', 'Balanced Accuracy', 'Precision', 'Recall', 'AUC'])
        

        calibration_df = pd.DataFrame(calibration_results, 
                                    columns=['Outcome', 'Slope', 'Intercept'])
        
        return metrics_df, calibration_df, calibration_plots
        
    except Exception as e:
        return f"Error processing data: {str(e)}", None, None

def create_interface():

    

    load_training_data()
    
    with gr.Blocks(
        css="""
        .gradio-container {
            max-width: none !important;
            height: 100vh;
            overflow-y: auto;
        }
        .main-container {
            padding: 20px;
        }
        .big-title {
            font-size: 2.5em;
            font-weight: bold;
            margin-bottom: 30px;
            text-align: center;
        }
        .section-title {
            font-size: 2em;
            font-weight: bold;
            margin: 40px 0 20px 0;
            color: #2d5aa0;
        }
        .subsection-title {
            font-size: 1.5em;
            font-weight: bold;
            margin: 30px 0 15px 0;
            color: #4a4a4a;
        }
        """,
        title="ML Model Evaluation Pipeline"
    ) as demo:
        
        with gr.Column(elem_classes=["main-container"]):

            gr.HTML('<div class="big-title">Input</div>')
            
            gr.Markdown("### Please upload the dataset:")
            file_input = gr.File(
                label="Upload Dataset (CSV)",
                file_types=[".csv"],
                type="filepath"
            )
            

            process_btn = gr.Button("Process Dataset", variant="primary", size="lg")
            

            gr.HTML('<div class="section-title">Outputs</div>')
            

            gr.HTML('<div class="subsection-title">Metrics</div>')
            metrics_table = gr.Dataframe(
                headers=["Outcome", "Accuracy", "Balanced Accuracy", "Precision", "Recall", "AUC"],
                interactive=False,
                wrap=True
            )
            

            gr.HTML('<div class="subsection-title">Calibration</div>')
            calibration_table = gr.Dataframe(
                headers=["Outcome", "Slope", "Intercept"],
                interactive=False,
                wrap=True
            )
            

            gr.Markdown("#### Calibration Curves")
            

            #plot1 = gr.Plot(label="Event Free Survival")
            plot2 = gr.Plot(label="Overall Survival") 
            plot3 = gr.Plot(label="Graft Failure")
            plot4 = gr.Plot(label="Acute GVHD")
            plot5 = gr.Plot(label="Chronic GVHD")
            plot6 = gr.Plot(label="Vaso-Occlusive Crisis Post-HCT")
            plot7 = gr.Plot(label="Stroke Post-HCT")
            
            plots = [plot2, plot3, plot4, plot5, plot6, plot7]
            

            def process_and_display(file):
                metrics_df, calibration_df, calibration_plots = train_and_evaluate(file)
                
                if isinstance(metrics_df, str):  # Error case
                    return metrics_df, None, None, None, None, None, None, None  # FIX 2: 8 values
                

                plot_outputs = [None] * 6
                if calibration_plots:
                    for i, plot in enumerate(calibration_plots[:6]):
                        plot_outputs[i] = plot
                
                return (metrics_df, calibration_df, 
                       plot_outputs[0], plot_outputs[1], plot_outputs[2], 
                       plot_outputs[3], plot_outputs[4], plot_outputs[5])
            

            process_btn.click(
                fn=process_and_display,
                inputs=[file_input],
                outputs=[metrics_table, calibration_table] + plots
            )
    
    return demo


if __name__ == "__main__":
    demo = create_interface()
    demo.launch(
        share=True,
        inbrowser=True,
        height=800,
        show_error=True
    )