Refactor settings.py: Replace Phi-2 model configuration with Neural Mistral 7B, enhancing reasoning and instruction following capabilities

Refactor settings.py: Update Llama 7B model configuration to TinyLlama 1.1B and replace OpenChat 3.5 with Microsoft Phi-2, enhancing performance details and documentation

Refactor settings.py: Update Mistral 7B Instruct model configuration for improved performance and documentation

Refactor settings.py: Update Llama 7B model configuration to TinyLlama 1.1B, enhancing performance details and documentation

Refactor settings.py: Update Mistral 7B Instruct model configuration for optimized CPU inference and improved documentation

Refactor settings.py: Remove Yi 6B Chat model configuration and update OpenChat 3.5 model ID for consistency

Refactor settings.py: Add Gemma 7B Instruct model configuration, enhancing capabilities and documentation

Refactor settings.py: Add OpenChat 3.5 model configuration, enhancing capabilities and documentation

Refactor settings.py: Remove GPT-3.5 Turbo and Claude-3-Haiku model configurations to streamline model management

Refactor settings.py: Add configurations for GPT-3.5 Turbo and Claude-3-Haiku models, enhancing capabilities and documentation

Refactor settings.py: Update model configurations to replace OpenAssistant SFT-1 and SOLAR 10.7B with Yi 6B Chat and Mixtral 8x7B, enhancing descriptions and capabilities

Refactor settings.py: Update model configurations to replace BLOOMZ 7B MT and mGPT with OpenAssistant SFT-1 and SOLAR 10.7B Instruct, enhancing descriptions and capabilities

Refactor settings.py: Update model configurations to include BLOOMZ 7B MT and mGPT with enhanced descriptions and capabilities

Refactor app.py and settings.py: Enhance Hugging Face token handling with improved validation and logging

Refactor app.py: Remove PRO_ONLY_MODELS and simplify model info update logic

Refactor settings.py: Simplify API configuration for free tier and remove account type checks

Refactor settings.py: Change default model from 'llama-7b' to 'zephyr-7b'

Refactor settings.py: Enhance API token retrieval with logging and improved error handling

Refactor settings.py: Load environment variables for API token retrieval

Refactor settings.py: Update token retrieval to use 'pro_api_token_2' and improve error message for missing token

Refactor settings.py: Update token retrieval to use 'pro_api_token' and improve error message for missing token

Refactor app.py: Ensure IS_PRO_ACCOUNT is available and improve comments for clarity

Refactor app.py: Add account status check and warning for PRO-only models in model info update

Refactor settings.py: Implement account type check for HuggingFace API and adjust settings based on account status.

Refactor app.py: Enhance error handling for API requests by adding a specific message for 403 Forbidden errors.

Refactor app.py: Remove redundant button click handlers for automatic knowledge base updates in the UI.

Refactor app.py: Improve layout and organization of Knowledge Base management UI, enhance button functionality, and streamline information retrieval process.

Refactor app.py: Remove unused functions for URL selection and knowledge base updates, and clean up event handlers in the Knowledge Base tab section.

Enhance README.md: Update knowledge base management section with new features, including custom URL handling, selective source management, and detailed update operations.

Refactor dataset.py: Update import path for HuggingFaceEmbeddings, streamline DatasetManager initialization, and enhance download_vector_store method with improved error handling and logging.

Refactor dataset.py: Replace print statements with logger.debug for better logging during upload checks and add error logging for upload failures.

Enhance knowledge base rebuilding functionality: Add method to rebuild knowledge base from selected URLs, improve error handling, and log actions during vector store download process.

Refactor app.py: Clean up imports by removing unused ones and correcting the import path for ChatEvaluator.

Enhance knowledge base update functionality: Add methods to retrieve selected URLs and update the knowledge base with selected sources, improving error handling and logging.

Refactor knowledge base update and vector store download: Simplify URL selection, enhance error handling, and improve temporary directory management in download_vector_store method.

Refactor app.py and dataset.py: Update comments for clarity, enhance download_vector_store method with improved error handling, and streamline vector store download process.

Enhance knowledge base management: Add functions to retrieve and save knowledge base metadata, improve error handling, and update constants for better clarity and functionality.

Refactor README and app.py: Update dataset structure in README, add knowledge base management features in app.py, and implement last update date retrieval in DatasetManager.

Enhance training method in FineTuner: Add detailed logging for training process, dataset loading, tokenization, and error handling to improve debugging and traceability.

Refactor FineTuner class: Remove commented-out code for model directory creation and the finetune_from_annotations function, clean up comments, and enhance code readability.

Refactor training method in FineTuner: Update training_data_path to be required, enhance gradient accumulation steps, and improve documentation for training parameters.

Refactor imports in fine_tuner.py: Add missing import for datetime module.

Refactor finetune_from_chat_history: Remove unnecessary parameter from ChatEvaluator instantiation and clean up comments.

Refactor fine-tuning process: Update file paths to utilize dataset constants and improve training data export handling.

Refactor fine-tuning process: Update function to utilize evaluated chat history and improve error handling for training data preparation.

Refactor evaluation interface: Update label for evaluated pairs checkbox and improve documentation for QA pairs dataframe function.

Refactor conversation deletion logic: Simplify annotation file path handling by reducing the number of paths considered for deletion.

Refactor training data export function and add dataset structure repair tool

Refactor conversation deletion logic: Enhance file deletion process for conversations by supporting multiple annotation file paths and updating file retrieval method.

Refactor conversation deletion logic: Update file retrieval method and enhance error handling for dataset access.