The evaluation-2023 from physionetchallenges

evaluation-2023's People

Contributors

Stargazers

Watchers

evaluation-2023's Issues

Some possible errors in evaluate_model.py

Should the following block

evaluation-2023/evaluate_model.py

Lines 130 to 150 in 3e69cdc

    
           # Compute the FPRs. 
        
           fpr = np.zeros(num_thresholds) 
        
           for l in range(num_thresholds): 
        
               if tp[l] + fn[l] > 0: 
        
                   fpr[l] = float(fp[l]) / float(tp[l] + fn[l]) 
        
               else: 
        
                   fpr[l] = float('nan') 
        
           # Find the threshold such that FPR <= 0.05. 
        
           max_fpr = 0.05 
        
           if np.any(fpr <= max_fpr): 
        
               l = max(l for l, x in enumerate(fpr) if x <= max_fpr) 
        
               tps[i] = tp[l] 
        
               fps[i] = fp[l] 
        
               fns[i] = fn[l] 
        
               tns[i] = tn[l] 
        
           else: 
        
               tps[i] = tp[0] 
        
               fps[i] = fp[0] 
        
               fns[i] = fn[0] 
        
               tns[i] = tn[0]

be out of the large for loop?

evaluation-2023/evaluate_model.py

Lines 113 to 151 in 3e69cdc

    
           # Update the TPs, FPs, FNs, and TNs using the values at the previous threshold. 
        
           k = 0 
        
           for l in range(1, num_thresholds): 
        
               tp[l] = tp[l-1] 
        
               fp[l] = fp[l-1] 
        
               fn[l] = fn[l-1] 
        
               tn[l] = tn[l-1] 
        
               while k < num_instances and current_outputs[idx[k]] >= thresholds[l]: 
        
                   if current_labels[idx[k]] == 1: 
        
                       tp[l] += 1 
        
                       fn[l] -= 1 
        
                   else: 
        
                       fp[l] += 1 
        
                       tn[l] -= 1 
        
                   k += 1 
        
               # Compute the FPRs. 
        
               fpr = np.zeros(num_thresholds) 
        
               for l in range(num_thresholds): 
        
                   if tp[l] + fn[l] > 0: 
        
                       fpr[l] = float(fp[l]) / float(tp[l] + fn[l]) 
        
                   else: 
        
                       fpr[l] = float('nan') 
        
               # Find the threshold such that FPR <= 0.05. 
        
               max_fpr = 0.05 
        
               if np.any(fpr <= max_fpr): 
        
                   l = max(l for l, x in enumerate(fpr) if x <= max_fpr) 
        
                   tps[i] = tp[l] 
        
                   fps[i] = fp[l] 
        
                   fns[i] = fn[l] 
        
                   tns[i] = tn[l] 
        
               else: 
        
                   tps[i] = tp[0] 
        
                   fps[i] = fp[0] 
        
                   fns[i] = fn[0] 
        
                   tns[i] = tn[0]

Wrong FPR in evaluate_model.py

In line 82 and 83 the same denominator is used for TPR and FPR:
float(tp[j] + fn[j])

It should be in line 83:
fpr[j] = float(fp[j]) / float(fp[j] + tn[j])

instead of
fpr[j] = float(fp[j]) / float(tp[j] + fn[j])

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.

Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

TensorFlow

An Open Source Machine Learning Framework for Everyone

Django

The Web framework for perfectionists with deadlines.

Laravel

A PHP framework for web artisans

D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

web

Some thing interesting about web. New door for the world.

server

A server is a program made to process requests and deliver data to clients.

Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

Visualization

Some thing interesting about visualization, use data art

Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.

Microsoft

Open source projects and samples from Microsoft.

Google

Google ❤️ Open Source for everyone.

Alibaba

Alibaba Open Source for everyone

D3

Data-Driven Documents codes.

Tencent

China tencent open source team.

physionetchallenges / evaluation-2023 Goto Github PK