TensorFlow：以官方卷积网络教程为基础适配自定义汽车数据集

I'm currently working on building a convolutional neural network using TensorFlow, drawing from the official TensorFlow convnet tutorial. To make this project more hands-on, I’ve swapped out the default tutorial dataset for the Stanford Cars dataset.

I’ve written a reader.py file to handle loading the custom dataset, and so far it’s running smoothly. Here’s a partial snippet of the working code:

import tensorflow as tf
from os import listdir
from os.path import isfile, join, dirname, realpath

def my_input_fn():
    this_path = dirname(realpath(__file__))
    data_path = this_path + "/cars_tr..."

(Note: The code snippet is truncated as it was cut off in the original input.)

内容的提问来源于stack exchange，提问作者Satler