zenUnicorn

model.compile(optimizer='adam',
             loss = 'binary_crossentropy',
             metrics=['accuracy'])

callback = keras.callbacks.EarlyStopping(monitor='val_loss',
                                            patience=3,
                                            restore_best_weights=True)

zenUnicorn

#We can also log the decision tree classifier parameters.
decision_tree_params ={
     "n_estimators": {
         "type": "discrete",
         "values": [50, 100, 150, 200, 250, 300]
     },
     "criterion": {
         "type": "categorical",
         "values": ["gini", "entropy"]
     },

zenUnicorn

import os
from PIL import Image


def image_crop(row):
    input_dir = 'Images\SampleImages'
    output_dir = 'Images\cropped_Images'
    
    # open image
    im = Image.open(f"{input_dir}\{row['filename']}")

zenUnicorn

hyperparams = {
    "batch_size": 32,
    "epochs": 20,
    "num_nodes": 64,
    "activation": 'relu',
    "optimizer": 'adam',
}
experiment.log_parameters(hyperparams)

zenUnicorn

import pandas as pd
import torch
import torchvision
from torchvision.transforms import transforms
import numpy as np
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
import torch.nn as nn
import torch.nn.functional as F
import comet_ml

zenUnicorn

label_data = pd.read_csv("facial_keypoints.csv")
label_data.head(5)

label_data.shape
label_data.describe()
label_data.info()

label_data = label_data.dropna()
label_index = label_data.index

zenUnicorn

image_data = np.load("face_images.npz")
image_data = image_data["face_images"]
image_data = np.transpose(image_data,(2,0,1))

full_image = np.expand_dims(image_data[label_index],1)
full_image

zenUnicorn

experiment = comet_ml.Experiment(
    api_key="API KEY",
    project_name="face-landmarks-recognition", 
    workspace="WORKSPACE-NAME",
    log_code=True)
    
hyperparameters = {
    "batch_size": 32,
    "epochs": 30,
    "test_size": 0.1,

zenUnicorn

X_train,X_test,y_train,y_test = train_test_split(full_image,full_label,test_size = 0.1)

zenUnicorn

class FaceDataset(torch.utils.data.Dataset):
    def __init__(self,image,label):
        self.image = image
        self.label = label
        
    def __len__(self):
        return len(self.image)
    
    def __getitem__(self,idx):
        image = self.image[idx] / 255.