torch_tsne_plotting.py

path_to_save_check_points=r'C:../early_stopping/' + '/test1'
path_to_checkpoints=path_to_save_check_points




import torch
import torch.backends.cudnn as cudnn
import torchvision
import torchvision.transforms as transforms
import torchvision.models as models

import argparse
import os

import numpy as np
import pandas as pd
# from tsne import bh_sne
from sklearn.manifold import TSNE
import seaborn as sns
import matplotlib.pyplot as plt
import timm
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"

LEARNING_RATE = 5e-3 #0.005
WEIGHT_DECAY = 5e-4 # 0.0005


def gen_features(test_loader):
    model = timm.create_model('vgg16', pretrained=True,num_classes=3)

    model =  model.to(device=DEVICE,dtype=torch.float)
    optimizer = torch.optim.Adam(model.parameters(), lr=LEARNING_RATE, weight_decay=WEIGHT_DECAY)
    checkpoint = torch.load(path_to_checkpoints+".pth.tar",map_location=DEVICE)
    model.load_state_dict(checkpoint['state_dict'])
    optimizer.load_state_dict(checkpoint['optimizer'])
    model.eval()
    targets_list = []
    outputs_list = []

    with torch.no_grad():
        for idx, (inputs, targets,images_name) in enumerate(test_loader):
            inputs = inputs.to(DEVICE)
            targets = targets.to(DEVICE)
            targets_np = targets.data.cpu().numpy()

            outputs = model(inputs)
            outputs_np = outputs.data.cpu().numpy()
            
            targets_list.append(targets_np[:, np.newaxis])
            outputs_list.append(outputs_np)
            
            if ((idx+1) % 10 == 0) or (idx+1 == len(test_loader)):
                print(idx+1, '/', len(test_loader))

    targets = np.concatenate(targets_list, axis=0)
    outputs = np.concatenate(outputs_list, axis=0).astype(np.float64)

    return targets, outputs


parser = argparse.ArgumentParser(description='PyTorch t-SNE for STL10')
parser.add_argument('--save-dir', type=str, default='./results', help='path to save the t-sne image')
parser.add_argument('--batch-size', type=int, default=12, help='batch size (default: 128)')
parser.add_argument('--seed', type=int, default=1, help='random seed value (default: 1)')

args = parser.parse_args()

if not os.path.exists(args.save_dir):
    os.makedirs(args.save_dir)


def tsne_plot(save_dir, targets, outputs):
    print('generating t-SNE plot...')
    # tsne_output = bh_sne(outputs)
    tsne = TSNE(random_state=0)
    tsne_output = tsne.fit_transform(outputs)

    df = pd.DataFrame(tsne_output, columns=['x', 'y'])
    df['targets'] = targets

    plt.rcParams['figure.figsize'] = 10, 10
    sns.scatterplot(
        x='x', y='y',
        hue='targets',
        palette=sns.color_palette("hls", 3), # change the number according to your outputs
        data=df,
        marker='o',
        legend="full",
        alpha=0.5
    )

    plt.xticks([])
    plt.yticks([])
    plt.xlabel('')
    plt.ylabel('')

    plt.savefig(os.path.join(save_dir,'tsne.png'), bbox_inches='tight')
    print('done!')

### need to pass the test data loader here 
targets, outputs = gen_features(test_loader)
tsne_plot(args.save_dir, targets, outputs)