1.Tested OS environments:
	Ubuntu 20.04
	Ubuntu 22.04
	
2. Tested Python version:
	Python 3.11.4
	
3. Required hardware configuration:
	If you want to try deep learning, it is necessary to have a CUDA-enabled device. Also, it is recommended to have a GPU memory above 10 Gb.
	
4. Required packages:

	1.1. Nessesary packages:

		pip install openai pandas thop imblearn lightning timm
		
		pip install scikit-learn==1.1.3
		
		conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia
						
		conda install -c conda-forge charset-normalizer
		
	
	1.2. If you want to try tasks for tabular data:

		pip install xgboost catboost lightgbm 
		
	1.2. If you want to try CV:

		pip install efficientnet-pytorch opencv-python scikit-image
		
		pip install albumentations==0.4.6

	1.3. If you want to try NLP:

		pip install nltk transformers sentencepiece spacy sacremoses

5. How to run an example
	
	First, type in your OpenAI API key in "key" in configs.json
	
	Then, cd path_to_the_folder_codes

	Then, python3 core.py
	

6. How to configure Text-to-ML:
	1. For the file config.json:
		Describe your datasets and task objectives by filling in "input data", "output data", "task objective", "evaluation metrics", and "files". Type in your openAI key in "key".
		GPT version: a string. Either "3.5", "4", or "4o"
		workspace: the path to the directory where the generated codes and the dataset are located. The format of the path should begin without a slash and end without a slash
	2. Ensure a stable internet connection for accessing OpenAI servers