Package index • aiworkflow

All functions

add_api_key_header(): Adds API key header to the qdrant connection for security

add_context(): Adds context to be used by the model when answering

add_tools_declaration(): Add tools declaration for the LLM to use

add_vision_capability(): Add vision capability to the workflow.

add_workflow_step(): Add a step (i.e. another workflow) to an existing workflow

ai_workflow(): Define AI workflow

apply_processing_skill(): Applies a processing skill to the current workflow

convert_batch_documents_to_embeddings(): Convert Batch documents to Embeddings

convert_embeddings_to_qdrant_format(): Convert embeddings to Qdrant format

convert_ollama_completion_response_to_tibble(): Convert an ollama server completion response to a tibble

convert_ollama_model_info_response_to_tibble(): Convert ollama response for model info to a tibble

convert_ollama_tags_response_to_tibble(): Convert an ollama server tags response to a tibble

create_custom_processing_skill(): Create a processing skill template file that can be used to create your own skills

display_intermediate_answer(): Display Intermediate Answer

execute_workflow(): Execute an AI workflow

execute_workflow_on_df(): Execute an AI workflow on a dataframe (with or without a pipe)

extract_snippets(): Extract Snippets

generate_document_embeddings(): Get embeddings for a piece of context through an ollama server instance020

generate_numeric_list(): Generate Numeric List

generate_uuid_from_text(): Generate UUID from text

get_list_ollama_models(): Get a list of models available from the ollama server

get_ollama_chat_completion(): Get chat completion from ollama server

get_ollama_completion(): Get a completion from ollama server

get_ollama_connection(): Define a connection to a local ollama server

get_ollama_embeddings(): Get embeddings for a piece of context through an ollama server instance

get_ollama_model_info(): Get information about one ollama model

get_qdrant_connection(): Get Qdrant connection

inspect_processing_skill(): Inspect a specific processing skill

list_global_functions(): List global functions

list_processing_skill_parameters(): List extra parameters for a given processing skill

list_processing_skills(): list the processing skills

load_context_embeddings_from_feather_files(): Load Context Embeddings From Feather Files

load_workflow(): Load workflow

make_cosine_similarity_matrix(): Make Cosine Similarity Matrix

parse_json_result(): Parse JSON answer from the LLM

process_prompts(): Process Prompts starting from a workflow

pull_final_answer(): Pull Final Answer

qdrant_check_collection_existence(): Qdrant: Check collection existence

qdrant_check_connection_validity(): Qdrant: Check if the Connection is valid

qdrant_create_new_collection(): Qdrant: Create new collection

qdrant_delete_collection(): Qdrant: Delete collection

qdrant_delete_points(): Qdrant: Delete points (vectors)

qdrant_get_collection_details(): Qdrant: Get collection details

qdrant_list_all_collections(): Qdrant: List all collections

qdrant_retrieve_point(): Qdrant: Retrieve a specific point (vector)

qdrant_search_points(): Qdrant: Search points (vectors)

qdrant_upsert_points(): Qdrant: Upsert points

request_json_answer(): Request JSON answer from the LLM

retrieve_similar_vectors(): Retrieve Similar Vectors

save_workflow(): Save workflow

set_audience(): Define a specific audience you want the model to prepare an answer for

set_connector(): Set the connector required to operate the workflow.

set_current_time_and_date_reference(): Set the current time and date as addition reference

set_custom_processing_skill(): Set a custom processing skill (that you created) to give to the workflow.

set_default_missing_parameters_in_workflow(): Set Defaults for missing workflow parameters

set_embedding_model(): Set the embedding model to be used by the workflow

set_frequency_penalty(): Set the frequency penalty of the model used by the flow.

set_ip_addr(): Set the IP Address required to connect to an API server.

set_mode(): Set the mode of the model used by the workflow.

set_model(): Set the LLM model to be used by the workflow

set_n_predict(): Set the number of tokens to be predicted (maximum) by the flow.

set_num_ctx(): Set the length of the context to be handled by the model

set_overall_background(): Set overall background info for your model before an answer is formulated

set_port(): Set the port required to connect to the API server.

set_presence_penalty(): Set the presence penalty of the model used by the flow.

set_processing_skill(): Set the processing skill that you want to give the workflow.

set_repeat_penalty(): Set the repeat penalty of the model used by the flow.

set_seed(): Set the seed of the model used by the workflow.

set_style_of_voice(): Define a specific style of voice that you want the LLM to use when answering

set_system_prompt(): Set the system prompt to be used by the model.

set_temperature(): Set the temperature of the model used by the workflow.

split_text_as_paragraphs(): Split text into paragraphs

split_text_as_sentences(): Split text into sentences

switch_to_workflow(): Switch to workflow

test_llamacpp_connection(): Confirm connection to a Llama.cpp server is working

test_ollama_connection(): Confirm connection to ollama is working

write_vectors_to_feather_file(): Write Vectors to Feather File