ChatGPT Chemistry Assistant
Step-by-step illustrations of setting up Processes 1, 2, and 3 were shown in the Supporting Information file of this article in the cookbook style.
Thank you!
Contents
· Text Mining: PDF Text Processing and Analysis with OpenAI's gpt-3.5-turbo API or gpt-4 API
· MOF Chatbot: a chatbot answers question based on post text mining data
· Predictive Model: A RF classfifier trained on post text mining data
Features
This text mining assistant includes the following main functions:
· Extraction of text from PDF files and its division into smaller chunks.
· Classfication of text segments.
· Processing and summarization of the extracted text data.
· Conversion of summarized data into a tabular format.
· Calculation of text embeddings using the OpenAI API.
· Selection of top similarity sections and their neighbors in the data.
· Calculation of text token count using the tiktoken library.
This MOF Synthesis Assistant tool provides the following core functionalities:
· Extraction of synthesis information and embeddings from a CSV file.
· Calculation of similarity scores.
· Sorting of text segments based on their similarity scores.
· Selection of top similar synthesis conditions from the sorted data.
· Processing of multiple user questions to maintain a conversational context.
· Use of the OpenAI API to generate text embeddings for user's questions based on the selected synthesis conditions.
· Maintenance of a conversation history for better contextually accurate responses in a conversational interface.
· A user-friendly conversational interface for asking questions related to MOF synthesis conditions.
This machine learning tool includes the following primary functions:
· Data Preprocessing: Reads, processes, and drops unused data columns from CSV file.
· Feature Selection: Applies RFECV for robust feature selection.
· Data Splitting: Splits data into training and testing sets with various sizes.
· Hyperparameter Tuning: Performs tuning via RandomizedSearchCV for RandomForestClassifier.
· Model Evaluation: Computes several performance metrics for each model configuration.
· Optimal Model Selection: Selects the best performing model based on balanced accuracy.
· Random Splits: Supports multiple random states for data splitting.
· Reporting: Records all performance metrics in an organized format for model comparison.
Dependencies
· This project is built on Python and requires the following libraries:
openai
requests
PyPDF2
pandas
tiktoken
sklearn
numpy
mendeleev