diff --git a/Project.toml b/Project.toml index dcef6d5..601d42d 100644 --- a/Project.toml +++ b/Project.toml @@ -1,7 +1,7 @@ name = "YiemAgent" uuid = "e012c34b-7f78-48e0-971c-7abb83b6f0a2" authors = ["narawat lamaiin "] -version = "0.1.2" +version = "0.1.3" [deps] DataFrames = "a93c6f00-e57d-5684-b7b6-d8193f3e46c0" diff --git a/src/interface.jl b/src/interface.jl index 7e3ae51..e44aafd 100644 --- a/src/interface.jl +++ b/src/interface.jl @@ -1,6 +1,6 @@ module interface -export addNewMessage, conversation, decisionMaker, evaluator, reflector, generatechat, +export addNewMessage, conversation, decisionMaker, reflector, generatechat, generalconversation, detectWineryName, generateSituationReport using JSON3, DataStructures, Dates, UUIDs, HTTP, Random, PrettyPrinting, Serialization, @@ -55,6 +55,8 @@ end config - `state::T2` a game state + +# Keyword Arguments # Return - `thoughtDict::Dict` @@ -90,8 +92,6 @@ julia> output_thoughtDict = Dict( # TODO - [] update docstring - - [x] implement the function - - [] implement RAG to pull similar experience - [] use customerinfo - [] user storeinfo @@ -294,15 +294,13 @@ function decisionMaker(a::T; recent::Integer=5)::Dict{Symbol,Any} where {T<:agen Dict(:name => "user", :text => usermsg) ] - # put in model format - prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct") - prompt *= """ - <|start_header_id|>assistant<|end_header_id|> - """ + # change qwen format put in model format + prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen") response = a.func[:text2textInstructLLM](prompt) response = GeneralUtils.remove_french_accents(response) - response = replace(response, '*'=>"") + response = replace(response, "**"=>"") + response = replace(response, "***"=>"") response = replace(response, "<|eot_id|>"=>"") # check if response contain more than one functions from ["CHATBOX", "CHECKINVENTORY", "ENDCONVERSATION"] @@ -318,9 +316,10 @@ function decisionMaker(a::T; recent::Integer=5)::Dict{Symbol,Any} where {T<:agen continue end - responsedict = GeneralUtils.textToDict(response, - ["Understanding", "Reasoning", "Plan", "Action_name", "Action_input"], - rightmarker=":", symbolkey=true, lowercasekey=true) + header = ["Understanding:", "Reasoning:", "Plan:", "Action_name:", "Action_input:"] + dictkey = ["understanding", "reasoning", "plan", "action_name", "action_input"] + responsedict = GeneralUtils.textToDict(response, header; + dictKey=dictkey, symbolkey=true) if responsedict[:action_name] ∉ ["CHATBOX", "CHECKINVENTORY", "ENDCONVERSATION"] errornote = "You must use the given functions" @@ -394,284 +393,280 @@ function decisionMaker(a::T; recent::Integer=5)::Dict{Symbol,Any} where {T<:agen end -""" Assigns a scalar value to each new child node to be used for selec- -tion and backpropagation. This value effectively quantifies the agent's progress in task completion, -serving as a heuristic to steer the search algorithm towards the most promising regions of the tree. +# """ Assigns a scalar value to each new child node to be used for selec- +# tion and backpropagation. This value effectively quantifies the agent's progress in task completion, +# serving as a heuristic to steer the search algorithm towards the most promising regions of the tree. -# Arguments - - `a::T1` - one of Yiem's agent - - `state::T2` - a game state +# # Arguments +# - `a::T1` +# one of Yiem's agent +# - `state::T2` +# a game state -# Return - - `evaluation::Tuple{String, Integer}` - evaluation and score +# # Return +# - `evaluation::Tuple{String, Integer}` +# evaluation and score -# Example -```jldoctest -julia> -``` +# # Example +# ```jldoctest +# julia> +# ``` -# Signature -""" -function evaluator(config::T1, state::T2 -)::Tuple{String,Integer} where {T1<:AbstractDict,T2<:AbstractDict} +# # Signature +# """ +# function evaluator(config::T1, state::T2 +# )::Tuple{String,Integer} where {T1<:AbstractDict,T2<:AbstractDict} - systemmsg = - """ - Analyze the trajectories of a solution to a question answering task. The trajectories are - labeled by environmental observations about the situation, thoughts that can reason about - the current situation and actions that can be three types: - 1) CHECKINVENTORY[query], which you can use to find wine in your inventory. - 2) CHATBOX[text], which you can use to interact with the user. +# systemmsg = +# """ +# Analyze the trajectories of a solution to a question answering task. The trajectories are +# labeled by environmental observations about the situation, thoughts that can reason about +# the current situation and actions that can be three types: +# 1) CHECKINVENTORY[query], which you can use to find wine in your inventory. +# 2) CHATBOX[text], which you can use to interact with the user. - Given a question and a trajectory, evaluate its correctness and provide your reasoning and - analysis in detail. Focus on the latest thought, action, and observation. Incomplete trajectories - can be correct if the thoughts and actions so far are correct, even if the answer is not found - yet. Do not generate additional thoughts or actions. Then ending with the correctness score s - where s is an integer from 0 to 10. +# Given a question and a trajectory, evaluate its correctness and provide your reasoning and +# analysis in detail. Focus on the latest thought, action, and observation. Incomplete trajectories +# can be correct if the thoughts and actions so far are correct, even if the answer is not found +# yet. Do not generate additional thoughts or actions. Then ending with the correctness score s +# where s is an integer from 0 to 10. - You should only respond in JSON format as describe below: - {"evaluation": "your evaluation", "score": "your evaluation score"} +# You should only respond in JSON format as describe below: +# {"evaluation": "your evaluation", "score": "your evaluation score"} - Here are some examples: - user: - { - "question": "I'm looking for a sedan with an automatic driving feature.", - "thought_1": "I have many types of sedans in my inventory, each with diverse features.", - "thought_2": "But there is only 1 model that has the feature customer wanted.", - "thought_3": "I should check our inventory first to see if we have it.", - "action_1": {"name": "inventory", "input": "Yiem model A"}, - "observation_1": "Yiem model A is in stock." - } - assistant - { - "evaluation": "This trajectory is correct as it is reasonable to check an inventory for info provided in the question. - It is also better to have simple searches corresponding to a single entity, making this the best action.", - "score": 10 - } +# Here are some examples: +# user: +# { +# "question": "I'm looking for a sedan with an automatic driving feature.", +# "thought_1": "I have many types of sedans in my inventory, each with diverse features.", +# "thought_2": "But there is only 1 model that has the feature customer wanted.", +# "thought_3": "I should check our inventory first to see if we have it.", +# "action_1": {"name": "inventory", "input": "Yiem model A"}, +# "observation_1": "Yiem model A is in stock." +# } +# assistant +# { +# "evaluation": "This trajectory is correct as it is reasonable to check an inventory for info provided in the question. +# It is also better to have simple searches corresponding to a single entity, making this the best action.", +# "score": 10 +# } - user: - { - "question": "Do you have an all-in-one pen with 4 colors and a pencil for sale?", - "thought_1": "Let me check our inventory first to see if I have it.", - "action_1": {"name": "inventory", "input": "pen with 4 color and a pencil."}, - "observation_1": "I found {1: "Pilot Dr. grip 4-in-1 pen", 2: "Rotting pencil"}", - "thought_2": "Ok, I have what the user is asking. Let's tell the user.", - "action_2": {"name": "CHATBOX", "input": "Yes, we do have a Pilot Dr. grip 4-in-1 pen and a Rotting pencil"}, - "observation_1": "This is not what I wanted." - } - assistant: - { - "evaluation": "This trajectory is incorrect as my search term should be related to a 4-colors pen with a pencil in it, - not a pen and a pencil seperately. A better search term should have been a 4-colors pen with a pencil, all-in-one.", - "score": 0 - } +# user: +# { +# "question": "Do you have an all-in-one pen with 4 colors and a pencil for sale?", +# "thought_1": "Let me check our inventory first to see if I have it.", +# "action_1": {"name": "inventory", "input": "pen with 4 color and a pencil."}, +# "observation_1": "I found {1: "Pilot Dr. grip 4-in-1 pen", 2: "Rotting pencil"}", +# "thought_2": "Ok, I have what the user is asking. Let's tell the user.", +# "action_2": {"name": "CHATBOX", "input": "Yes, we do have a Pilot Dr. grip 4-in-1 pen and a Rotting pencil"}, +# "observation_1": "This is not what I wanted." +# } +# assistant: +# { +# "evaluation": "This trajectory is incorrect as my search term should be related to a 4-colors pen with a pencil in it, +# not a pen and a pencil seperately. A better search term should have been a 4-colors pen with a pencil, all-in-one.", +# "score": 0 +# } - Let's begin! - """ +# Let's begin! +# """ - usermsg = """ - $(JSON3.write(state[:thoughtHistory])) - """ +# usermsg = """ +# $(JSON3.write(state[:thoughtHistory])) +# """ - chathistory = - [ - Dict(:name => "system", :text => systemmsg), - Dict(:name => "user", :text => usermsg) - ] +# chathistory = +# [ +# Dict(:name => "system", :text => systemmsg), +# Dict(:name => "user", :text => usermsg) +# ] - # put in model format - prompt = formatLLMtext(chathistory, "llama3instruct") - prompt *= """ - <|start_header_id|>assistant<|end_header_id|> - { - """ +# # put in model format +# prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen") - pprint(prompt) - externalService = config[:externalservice][:text2textinstruct] +# pprint(prompt) +# externalService = config[:externalservice][:text2textinstruct] - # apply LLM specific instruct format - externalService = config[:externalservice][:text2textinstruct] +# # apply LLM specific instruct format +# externalService = config[:externalservice][:text2textinstruct] - msgMeta = GeneralUtils.generate_msgMeta( - externalService[:mqtttopic], - senderName="evaluator", - senderId=string(uuid4()), - receiverName="text2textinstruct", - mqttBroker=config[:mqttServerInfo][:broker], - mqttBrokerPort=config[:mqttServerInfo][:port], - ) +# msgMeta = GeneralUtils.generate_msgMeta( +# externalService[:mqtttopic], +# senderName="evaluator", +# senderId=string(uuid4()), +# receiverName="text2textinstruct", +# mqttBroker=config[:mqttServerInfo][:broker], +# mqttBrokerPort=config[:mqttServerInfo][:port], +# ) - outgoingMsg = Dict( - :msgMeta => msgMeta, - :payload => Dict( - :text => prompt, - :kwargs => Dict( - :max_tokens => 512, - :stop => ["<|eot_id|>"], - ) - ) - ) +# outgoingMsg = Dict( +# :msgMeta => msgMeta, +# :payload => Dict( +# :text => prompt, +# :kwargs => Dict( +# :max_tokens => 512, +# :stop => ["<|eot_id|>"], +# ) +# ) +# ) - for attempt in 1:5 - try - response = GeneralUtils.sendReceiveMqttMsg(outgoingMsg) - _responseJsonStr = response[:response][:text] - expectedJsonExample = """ - Here is an expected JSON format: - {"evaluation": "...", "score": "..."} - """ - responseJsonStr = jsoncorrection(config, _responseJsonStr, expectedJsonExample) - evaluationDict = copy(JSON3.read(responseJsonStr)) +# for attempt in 1:5 +# try +# response = GeneralUtils.sendReceiveMqttMsg(outgoingMsg) +# _responseJsonStr = response[:response][:text] +# expectedJsonExample = """ +# Here is an expected JSON format: +# {"evaluation": "...", "score": "..."} +# """ +# responseJsonStr = jsoncorrection(config, _responseJsonStr, expectedJsonExample) +# evaluationDict = copy(JSON3.read(responseJsonStr)) - # check if dict has all required value - dummya::AbstractString = evaluationDict[:evaluation] - dummyb::Integer = evaluationDict[:score] +# # check if dict has all required value +# dummya::AbstractString = evaluationDict[:evaluation] +# dummyb::Integer = evaluationDict[:score] - return (evaluationDict[:evaluation], evaluationDict[:score]) - catch e - io = IOBuffer() - showerror(io, e) - errorMsg = String(take!(io)) - st = sprint((io, v) -> show(io, "text/plain", v), stacktrace(catch_backtrace())) - println("\nAttempt $attempt. Error occurred: $errorMsg\n$st ", Dates.now(), " ", @__FILE__, " ", @__LINE__) - end - end - error("evaluator failed to generate an evaluation") -end +# return (evaluationDict[:evaluation], evaluationDict[:score]) +# catch e +# io = IOBuffer() +# showerror(io, e) +# errorMsg = String(take!(io)) +# st = sprint((io, v) -> show(io, "text/plain", v), stacktrace(catch_backtrace())) +# println("\nAttempt $attempt. Error occurred: $errorMsg\n$st ", Dates.now(), " ", @__FILE__, " ", @__LINE__) +# end +# end +# error("evaluator failed to generate an evaluation") +# end -""" +# """ -# Arguments +# # Arguments -# Return +# # Return -# Example -```jldoctest -julia> -``` +# # Example +# ```jldoctest +# julia> +# ``` -# TODO - - [] update docstring - - [x] implement the function - - [x] add try block. check result that it is expected before returning +# # TODO +# - [] update docstring +# - [x] implement the function +# - [x] add try block. check result that it is expected before returning -# Signature -""" -function reflector(config::T1, state::T2)::String where {T1<:AbstractDict,T2<:AbstractDict} - # https://github.com/andyz245/LanguageAgentTreeSearch/blob/main/hotpot/hotpot.py +# # Signature +# """ +# function reflector(config::T1, state::T2)::String where {T1<:AbstractDict,T2<:AbstractDict} +# # https://github.com/andyz245/LanguageAgentTreeSearch/blob/main/hotpot/hotpot.py - _prompt = - """ - You are a helpful sommelier working for a wine store. - Your goal is to recommend the best wine from your inventory that match the user preferences. - You will be given a question and a trajectory of the previous help you've done for a user. - You were unsuccessful in helping the user either because you guessed the wrong answer with Finish[answer], or you didn't know the user enough. - In a few sentences, Diagnose a possible reason for failure and devise a new, concise, high level plan that aims to mitigate the same failure. - Use complete sentences. +# _prompt = +# """ +# You are a helpful sommelier working for a wine store. +# Your goal is to recommend the best wine from your inventory that match the user preferences. +# You will be given a question and a trajectory of the previous help you've done for a user. +# You were unsuccessful in helping the user either because you guessed the wrong answer with Finish[answer], or you didn't know the user enough. +# In a few sentences, Diagnose a possible reason for failure and devise a new, concise, high level plan that aims to mitigate the same failure. +# Use complete sentences. - You should only respond in JSON format as describe below: - {"reflection": "your relection"} +# You should only respond in JSON format as describe below: +# {"reflection": "your relection"} - Here are some examples: - Previous Trial: - { - "question": "Hello, I would like a get a bottle of wine", - "thought_1": "A customer wants to buy a bottle of wine. Before making a recommendation, I need to know more about their preferences.", - "action_1": {"name": "CHATBOX", "input": "What is the occasion for which you're buying this wine?"}, - "observation_1": "We are holding a wedding party", +# Here are some examples: +# Previous Trial: +# { +# "question": "Hello, I would like a get a bottle of wine", +# "thought_1": "A customer wants to buy a bottle of wine. Before making a recommendation, I need to know more about their preferences.", +# "action_1": {"name": "CHATBOX", "input": "What is the occasion for which you're buying this wine?"}, +# "observation_1": "We are holding a wedding party", - "thought_2": "A wedding party, that's a great occasion! The customer might be looking for a celebratory drink. Let me ask some more questions to narrow down the options.", - "action_2": {"name": "CHATBOX", "input": "What type of food will you be serving at the wedding?"}, - "observation_2": "It will be Thai dishes.", +# "thought_2": "A wedding party, that's a great occasion! The customer might be looking for a celebratory drink. Let me ask some more questions to narrow down the options.", +# "action_2": {"name": "CHATBOX", "input": "What type of food will you be serving at the wedding?"}, +# "observation_2": "It will be Thai dishes.", - "thought_3": "With Thai food, I should recommend a wine that complements its spicy and savory flavors. And since it's a celebratory occasion, the customer might prefer a full-bodied wine.", - "action_3": {"name": "CHATBOX", "input": "What is your budget for this bottle of wine?"}, - "observation_3": "I would spend up to 50 bucks.", +# "thought_3": "With Thai food, I should recommend a wine that complements its spicy and savory flavors. And since it's a celebratory occasion, the customer might prefer a full-bodied wine.", +# "action_3": {"name": "CHATBOX", "input": "What is your budget for this bottle of wine?"}, +# "observation_3": "I would spend up to 50 bucks.", - "thought_4": "Now that I have some more information, it's time to narrow down the options.", - "action_4": {"name": "winestock", "input": "red wine with full body, pairs well with spicy food, budget \$50"}, - "observation_4": "I found the following wines in our stock: \n{\n 1: El Enemigo Cabernet Franc 2019\n2: Tantara Chardonnay 2017\n\n}\n", +# "thought_4": "Now that I have some more information, it's time to narrow down the options.", +# "action_4": {"name": "winestock", "input": "red wine with full body, pairs well with spicy food, budget \$50"}, +# "observation_4": "I found the following wines in our stock: \n{\n 1: El Enemigo Cabernet Franc 2019\n2: Tantara Chardonnay 2017\n\n}\n", - "thought_5": "Now that I have a list of potential wines, I need to know more about the customer's taste preferences.", - "action_5": {"name": "CHATBOX", "input": "What type of wine characteristics are you looking for? (e.g. tannin level, sweetness, intensity, acidity)"}, - "observation_5": "I like full-bodied red wine with low tannin.", +# "thought_5": "Now that I have a list of potential wines, I need to know more about the customer's taste preferences.", +# "action_5": {"name": "CHATBOX", "input": "What type of wine characteristics are you looking for? (e.g. tannin level, sweetness, intensity, acidity)"}, +# "observation_5": "I like full-bodied red wine with low tannin.", - "thought_6": "Now that I have more information about the customer's preferences, it's time to make a recommendation.", - "action_6": {"name": "recommendbox", "input": "El Enemigo Cabernet Franc 2019"}, - "observation_6": "I don't like the one you recommend. I want dry wine." - } +# "thought_6": "Now that I have more information about the customer's preferences, it's time to make a recommendation.", +# "action_6": {"name": "recommendbox", "input": "El Enemigo Cabernet Franc 2019"}, +# "observation_6": "I don't like the one you recommend. I want dry wine." +# } - { - "reflection": "I asked the user about the occasion, food type, and budget, and then searched for wine in the inventory right away. However, I should have asked the user for the specific wine type and their preferences in order to gather more information before making a recommendation." - } +# { +# "reflection": "I asked the user about the occasion, food type, and budget, and then searched for wine in the inventory right away. However, I should have asked the user for the specific wine type and their preferences in order to gather more information before making a recommendation." +# } - Let's begin! +# Let's begin! - Previous trial: - $(JSON3.write(state[:thoughtHistory])) - {"reflection" - """ +# Previous trial: +# $(JSON3.write(state[:thoughtHistory])) +# {"reflection" +# """ - # apply LLM specific instruct format - externalService = config[:externalservice][:text2textinstruct] - llminfo = externalService[:llminfo] - prompt = - if llminfo[:name] == "llama3instruct" - formatLLMtext_llama3instruct("system", _prompt) - else - error("llm model name is not defied yet $(@__LINE__)") - end +# # apply LLM specific instruct format +# externalService = config[:externalservice][:text2textinstruct] +# llminfo = externalService[:llminfo] +# prompt = +# if llminfo[:name] == "llama3instruct" +# formatLLMtext_llama3instruct("system", _prompt) +# else +# error("llm model name is not defied yet $(@__LINE__)") +# end - msgMeta = GeneralUtils.generate_msgMeta( - a.config[:externalservice][:text2textinstruct][:mqtttopic], - senderName="reflector", - senderId=string(uuid4()), - receiverName="text2textinstruct", - mqttBroker=config[:mqttServerInfo][:broker], - mqttBrokerPort=config[:mqttServerInfo][:port], - ) +# msgMeta = GeneralUtils.generate_msgMeta( +# a.config[:externalservice][:text2textinstruct][:mqtttopic], +# senderName="reflector", +# senderId=string(uuid4()), +# receiverName="text2textinstruct", +# mqttBroker=config[:mqttServerInfo][:broker], +# mqttBrokerPort=config[:mqttServerInfo][:port], +# ) - outgoingMsg = Dict( - :msgMeta => msgMeta, - :payload => Dict( - :text => prompt, - :kwargs => Dict( - :max_tokens => 512, - :stop => ["<|eot_id|>"], - ) - ) - ) +# outgoingMsg = Dict( +# :msgMeta => msgMeta, +# :payload => Dict( +# :text => prompt, +# :kwargs => Dict( +# :max_tokens => 512, +# :stop => ["<|eot_id|>"], +# ) +# ) +# ) - for attempt in 1:5 - try - response = GeneralUtils.sendReceiveMqttMsg(outgoingMsg) - _responseJsonStr = response[:response][:text] - expectedJsonExample = """ - Here is an expected JSON format: - {"reflection": "..."} - """ - responseJsonStr = jsoncorrection(config, _responseJsonStr, expectedJsonExample) - reflectionDict = copy(JSON3.read(responseJsonStr)) +# for attempt in 1:5 +# try +# response = GeneralUtils.sendReceiveMqttMsg(outgoingMsg) +# _responseJsonStr = response[:response][:text] +# expectedJsonExample = """ +# Here is an expected JSON format: +# {"reflection": "..."} +# """ +# responseJsonStr = jsoncorrection(config, _responseJsonStr, expectedJsonExample) +# reflectionDict = copy(JSON3.read(responseJsonStr)) - # check if dict has all required value - dummya::AbstractString = reflectionDict[:reflection] +# # check if dict has all required value +# dummya::AbstractString = reflectionDict[:reflection] - return reflectionDict[:reflection] - catch e - io = IOBuffer() - showerror(io, e) - errorMsg = String(take!(io)) - st = sprint((io, v) -> show(io, "text/plain", v), stacktrace(catch_backtrace())) - println("\nAttempt $attempt. Error occurred: $errorMsg\n$st ", Dates.now(), " ", @__FILE__, " ", @__LINE__) - end - end - error("reflector failed to generate a thought") -end +# return reflectionDict[:reflection] +# catch e +# io = IOBuffer() +# showerror(io, e) +# errorMsg = String(take!(io)) +# st = sprint((io, v) -> show(io, "text/plain", v), stacktrace(catch_backtrace())) +# println("\nAttempt $attempt. Error occurred: $errorMsg\n$st ", Dates.now(), " ", @__FILE__, " ", @__LINE__) +# end +# end +# error("reflector failed to generate a thought") +# end """ Chat with llm. @@ -863,46 +858,6 @@ function think(a::T)::NamedTuple{(:actionname, :result),Tuple{String,String}} wh ) result = chatresponse - # # store thoughtDict after the conversation finish - # if a.memory[:events][end][:thought][:action_name] == "ENDCONVERSATION" - # # generateSituationReport in the agent didn't include the last conversation - # # so the function will be called here - # a.memory[:recap] = generateSituationReport(a, a.func[:text2textInstructLLM]; skiprecent=0) - - # for (i, event) in enumerate(a.memory[:events]) - # if event[:subject] == "assistant" - # # create timeline of the last 3 conversation except the last one. - # # The former will be used as caching key and the latter will be the caching target - # # in vector database - # all_recapkeys = keys(a.memory[:recap]) # recap as caching - # all_recapkeys_vec = [r for r in all_recapkeys] # convert to a vector - - # # select from 1 to 2nd-to-lase event (i.e. excluding the latest which is assistant's response) - # _recapkeys_vec = all_recapkeys_vec[1:i-1] - - # # select only previous 3 recaps - # recapkeys_vec = - # if length(_recapkeys_vec) <= 3 # 1st message is a user's hello msg - # _recapkeys_vec # choose all - # else - # _recapkeys_vec[end-2:end] - # end - # #[PENDING] if there is specific data such as number, donot store in database - # tempmem = DataStructures.OrderedDict() - # for k in recapkeys_vec - # tempmem[k] = a.memory[:recap][k] - # end - - # recap = GeneralUtils.dictToString_noKey(tempmem) - # thoughtDict = a.memory[:events][i][:thought] # latest assistant thoughtDict - # a.func[:insertSommelierDecision](recap, thoughtDict) - # else - # # skip - # end - # end - # println("Caching conversation done") - # end - elseif actionname == "CHECKINVENTORY" if rawresponse !== nothing vd = GeneralUtils.dfToVectorDict(rawresponse) @@ -999,6 +954,9 @@ function generatechat(a::sommelier, thoughtDict) Let's begin! """ + header = ["Chat:"] + dictkey = ["chat"] + # a.memory[:shortmem][:available_wine] is a vector of dictionary context = if length(a.memory[:shortmem][:available_wine]) != 0 @@ -1036,10 +994,7 @@ function generatechat(a::sommelier, thoughtDict) ] # put in model format - prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct") - prompt *= """ - <|start_header_id|>assistant<|end_header_id|> - """ + prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen") try response = a.func[:text2textInstructLLM](prompt) @@ -1054,22 +1009,27 @@ function generatechat(a::sommelier, thoughtDict) response = replace(response, '`' => "") response = replace(response, "<|eot_id|>"=>"") response = GeneralUtils.remove_french_accents(response) - responsedict = GeneralUtils.textToDict(response, ["Chat"], - rightmarker=":", symbolkey=true, lowercasekey=true) - for i ∈ [:chat] - if length(JSON3.write(responsedict[i])) == 0 - error("$i is empty ", Dates.now(), " ", @__FILE__, " ", @__LINE__) - end + # check whether response has all header + detected_kw = GeneralUtils.detect_keyword(header, response) + if sum(values(detected_kw)) < length(header) + errornote = "\nSQL decisionMaker() response does not have all header" + continue + elseif sum(values(detected_kw)) > length(header) + errornote = "\nSQL decisionMaker() response has duplicated header" + continue end - # check if there are more than 1 key per categories - for i ∈ [:chat] - matchkeys = GeneralUtils.findMatchingDictKey(responsedict, i) - if length(matchkeys) > 1 - error("generatechat has more than one key per categories") - end - end + responsedict = GeneralUtils.textToDict(response, header; + dictKey=dictkey, symbolkey=true) + + # # check if there are more than 1 key per categories + # for i ∈ Symbol.(dictkey) + # matchkeys = GeneralUtils.findMatchingDictKey(responsedict, i) + # if length(matchkeys) > 1 + # error("generatechat has more than one key per categories") + # end + # end # check if Context: is in chat if occursin("Context:", responsedict[:chat]) @@ -1161,10 +1121,7 @@ function generatechat(a::companion) ] # put in model format - prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct") - prompt *= """ - <|start_header_id|>assistant<|end_header_id|> - """ + prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen") response = a.text2textInstructLLM(prompt) @@ -1272,7 +1229,7 @@ function generatequestion(a, text2textInstructLLM::Function; recent=nothing)::St Let's begin! """ - + context = if length(a.memory[:shortmem][:available_wine]) != 0 "Available wines you've found in your inventory so far: $(availableWineToText(a.memory[:shortmem][:available_wine]))" @@ -1333,13 +1290,36 @@ function generatequestion(a, text2textInstructLLM::Function; recent=nothing)::St ] # put in model format - prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct") - prompt *= """ - <|start_header_id|>assistant<|end_header_id|> - """ + prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen") try response = text2textInstructLLM(prompt) + # make sure generatequestion() don't have wine name that is not from retailer inventory + # check whether an agent recommend wines before checking inventory or recommend wines + # outside its inventory + # ask LLM whether there are any winery mentioned in the response + mentioned_winery = detectWineryName(a, response) + if mentioned_winery != "None" + mentioned_winery = String.(strip.(split(mentioned_winery, ","))) + + # check whether the wine is in event + isWineInEvent = false + for winename in mentioned_winery + for event in a.memory[:events] + if event[:outcome] !== nothing && occursin(winename, event[:outcome]) + isWineInEvent = true + break + end + end + end + + # if wine is mentioned but not in timeline or shortmem, + # then the agent is not supposed to recommend the wine + if isWineInEvent == false + errornote = "Previously, You mentioned wines that is not in your inventory which is not allowed." + error("Previously, You mentioned wines that is not in your inventory which is not allowed.") + end + end # sometime LLM generate more than 1 Understanding: understanding_number = count("Understanding:", response) @@ -1359,9 +1339,10 @@ function generatequestion(a, text2textInstructLLM::Function; recent=nothing)::St error("no answer found in the response ", Dates.now(), " ", @__FILE__, " ", @__LINE__) end - responsedict = GeneralUtils.textToDict(response, - ["Understanding", "Q1"], - rightmarker=":", symbolkey=true, lowercasekey=true) + header = ["Understanding:", "Q1:"] + dictkey = ["understanding", "q1"] + responsedict = GeneralUtils.textToDict(response, header; + dictKey=dictkey, symbolkey=true) response = "Q1: " * responsedict[:q1] println("\n~~~ generatequestion ", Dates.now(), " ", @__FILE__, " ", @__LINE__) pprintln(response) @@ -1412,6 +1393,9 @@ function generateSituationReport(a, text2textInstructLLM::Function; skiprecent:: Let's begin! """ + header = ["Event_$i:" for i in eachindex(a.memory[:events])] + dictkey = lowercase.(["Event_$i" for i in eachindex(a.memory[:events])]) + if length(a.memory[:events]) <= skiprecent return nothing end @@ -1437,15 +1421,11 @@ function generateSituationReport(a, text2textInstructLLM::Function; skiprecent:: ] # put in model format - prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct") - prompt *= """ - <|start_header_id|>assistant<|end_header_id|> - """ + prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen") response = text2textInstructLLM(prompt) - eventheader = ["Event_$i" for i in eachindex(a.memory[:events])] - responsedict = GeneralUtils.textToDict(response, eventheader, - rightmarker=":", symbolkey=true) + responsedict = GeneralUtils.textToDict(response, header; + dictKey=dictkey, symbolkey=true) println("\n~~~ generateSituationReport() ", Dates.now(), " ", @__FILE__, " ", @__LINE__) pprintln(response) @@ -1494,18 +1474,17 @@ function detectWineryName(a, text) ] # put in model format - prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct") - prompt *= """ - <|start_header_id|>assistant<|end_header_id|> - """ + prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen") try response = a.func[:text2textInstructLLM](prompt) println("\n~~~ detectWineryName() ", Dates.now(), " ", @__FILE__, " ", @__LINE__) pprintln(response) - responsedict = GeneralUtils.textToDict(response, ["winery_names"], - rightmarker=":", symbolkey=true, lowercasekey=true) + header = ["Winery_names:"] + dictkey = ["winery_names"] + responsedict = GeneralUtils.textToDict(response, header; + dictKey=dictkey, symbolkey=true) result = responsedict[:winery_names] diff --git a/src/llmfunction.jl b/src/llmfunction.jl index cffc122..62c15ec 100644 --- a/src/llmfunction.jl +++ b/src/llmfunction.jl @@ -326,7 +326,7 @@ julia> # TODO - [] update docstring - - [x] implement the function + - implement the function # Signature """ @@ -336,31 +336,41 @@ function extractWineAttributes_1(a::T1, input::T2)::String where {T1<:agent, T2< """ As a helpful sommelier, your task is to extract the user information from the user's query as much as possible to fill out user's preference form. - At each round of conversation, the user will give you the current situation: + At each round of conversation, the user will give you the following: User's query: ... You must follow the following guidelines: - 1) If specific information required in the preference form is not available in the query or there isn't any, mark with "NA" to indicate this. + - If specific information required in the preference form is not available in the query or there isn't any, mark with "NA" to indicate this. Additionally, words like 'any' or 'unlimited' mean no information is available. - 2) Do not generate other comments. + - Do not generate other comments. - You should then respond to the user with the following points: - - reasoning: state your understanding of the current situation - - wine_name: name of the wine - - winery: name of the winery - - vintage: the year of the wine - - region: a region (NOT a country) where the wine is produced, such as Burgundy, Napa Valley, etc - - country: a country where the wine is produced. Can be "Austria", "Australia", "France", "Germany", "Italy", "Portugal", "Spain", "United States" - - wine_type: can be one of: "red", "white", "sparkling", "rose", "dessert" or "fortified" - - grape_varietal: the name of the primary grape used to make the wine - - tasting_notes: a brief description of the wine's taste, such as "butter", "oak", "fruity", etc - - wine_price: price range of wine. - - occasion: the occasion the user is having the wine for - - food_to_be_paired_with_wine: food that the user will be served with the wine such as poultry, fish, steak, etc - + You should then respond to the user with: + Comprehension: state your understanding of the current situation + Wine_name: name of the wine + Winery: name of the winery + Vintage: the year of the wine + Region: a region (NOT a country) where the wine is produced, such as Burgundy, Napa Valley, etc + Country: a country where the wine is produced. Can be "Austria", "Australia", "France", "Germany", "Italy", "Portugal", "Spain", "United States" + Wine_type: can be one of: "red", "white", "sparkling", "rose", "dessert" or "fortified" + Grape_varietal: the name of the primary grape used to make the wine + Tasting_notes: a brief description of the wine's taste, such as "butter", "oak", "fruity", etc + Wine_price: price range of wine. + Occasion: the occasion the user is having the wine for + Food_to_be_paired_with_wine: food that the user will be served with the wine such as poultry, fish, steak, etc - You should only respond in the user's preference form (JSON) as described below: - {"reasoning": ..., "winery": ..., "wine_name": ..., "vintage": ..., "region": ..., "country": ..., "wine_type": ..., "grape_varietal": ..., "tasting_notes": ..., "wine_price": ..., "occasion": ..., "food_to_be_paired_with_wine": ...} + You should only respond in format as described below: + Comprehension: ... + Wine_name: ... + Winery: ... + Vintage: ... + Region: ... + Country: ... + Wine_type: + Grape_varietal: ... + Tasting_notes: ... + Wine_price: ... + Occasion: ... + Food_to_be_paired_with_wine: ... Here are some example: User's query: red, Chenin Blanc, Riesling, 20 USD @@ -372,7 +382,8 @@ function extractWineAttributes_1(a::T1, input::T2)::String where {T1<:agent, T2< Let's begin! """ - attributes = ["reasoning", "winery", "wine_name", "vintage", "region", "country", "wine_type", "grape_varietal", "tasting_notes", "wine_price", "occasion", "food_to_be_paired_with_wine"] + header = ["Comprehension:", "Wine_name:", "Winery:", "Vintage:", "Region:", "Country:", "Wine_type:", "Grape_varietal:", "Tasting_notes:", "Wine_price:", "Occasion:", "Food_to_be_paired_with_wine:"] + dictkey = ["comprehension", "wine_name", "winery", "vintage", "region", "country", "wine_type", "grape_varietal", "tasting_notes", "wine_price", "occasion", "food_to_be_paired_with_wine"] errornote = "" for attempt in 1:5 @@ -389,18 +400,13 @@ function extractWineAttributes_1(a::T1, input::T2)::String where {T1<:agent, T2< ] # put in model format - prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct") - prompt *= - """ - <|start_header_id|>assistant<|end_header_id|> - """ - + prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen") response = a.func[:text2textInstructLLM](prompt) response = GeneralUtils.remove_french_accents(response) # check wheter all attributes are in the response checkFlag = false - for word in attributes + for word in header if !occursin(word, response) errornote = "$word attribute is missing in previous attempts" println("Attempt $attempt $errornote ", Dates.now(), " ", @__FILE__, " ", @__LINE__) @@ -409,12 +415,20 @@ function extractWineAttributes_1(a::T1, input::T2)::String where {T1<:agent, T2< end end checkFlag == true ? continue : nothing + + # check whether response has all header + detected_kw = GeneralUtils.detect_keyword(header, response) + if sum(values(detected_kw)) < length(header) + errornote = "\nYiemAgent extractWineAttributes_1() response does not have all header" + continue + elseif sum(values(detected_kw)) > length(header) + errornote = "\nYiemAgent extractWineAttributes_1() response has duplicated header" + continue + end + responsedict = GeneralUtils.textToDict(response, header; + dictKey=dictkey, symbolkey=true) - responsedict = copy(JSON3.read(response)) - - # convert - - delete!(responsedict, :reasoning) + delete!(responsedict, :comprehension) delete!(responsedict, :tasting_notes) delete!(responsedict, :occasion) delete!(responsedict, :food_to_be_paired_with_wine) @@ -424,9 +438,9 @@ function extractWineAttributes_1(a::T1, input::T2)::String where {T1<:agent, T2< # check if winery, wine_name, region, country, wine_type, grape_varietal's value are in the query because sometime AI halucinates checkFlag = false - for i in attributes + for i in dictkey j = Symbol(i) - if j ∉ [:reasoning, :tasting_notes, :occasion, :food_to_be_paired_with_wine] + if j ∉ [:comprehension, :tasting_notes, :occasion, :food_to_be_paired_with_wine] # in case j is wine_price it needs to be checked differently because its value is ranged if j == :wine_price if responsedict[:wine_price] != "NA" @@ -509,7 +523,7 @@ function extractWineAttributes_2(a::T1, input::T2)::String where {T1<:agent, T2< conversiontable = """ - Conversion Table: + Intensity level: 1 to 2: May correspond to "light-bodied" or a similar description. 2 to 3: May correspond to "med light bodied", "medium light" or a similar description. @@ -534,6 +548,7 @@ function extractWineAttributes_2(a::T1, input::T2)::String where {T1<:agent, T2< 3 to 4: May correspond to "medium acidity" or a similar description. 4 to 5: May correspond to "semi high acidity" or a similar description. 4 to 5: May correspond to "high acidity" or a similar description. + """ systemmsg = @@ -547,67 +562,64 @@ function extractWineAttributes_2(a::T1, input::T2)::String where {T1<:agent, T2< The preference form requires the following information: sweetness, acidity, tannin, intensity - You must follow the following guidelines: + 1) If specific information required in the preference form is not available in the query or there isn't any, mark with 'NA' to indicate this. Additionally, words like 'any' or 'unlimited' mean no information is available. 2) Use the conversion table to convert the descriptive word level of sweetness, intensity, tannin, and acidity into a corresponding integer. 3) Do not generate other comments. + - You should then respond to the user with the following points: - - sweetness_keyword: The exact keywords in the user's query describing the sweetness level of the wine. - - sweetness: ( S ), where ( S ) represents integers indicating the range of sweetness levels. Example: 1-2 - - acidity_keyword: The exact keywords in the user's query describing the acidity level of the wine. - - acidity: ( A ), where ( A ) represents integers indicating the range of acidity level. Example: 3-5 - - tannin_keyword: The exact keywords in the user's query describing the tannin level of the wine. - - tannin: ( T ), where ( T ) represents integers indicating the range of tannin level. Example: 1-3 - - intensity_keyword: The exact keywords in the user's query describing the intensity level of the wine. - - intensity: ( I ), where ( I ) represents integers indicating the range of intensity level. Example: 2-4 + + Sweetness_keyword: The exact keywords in the user's query describing the sweetness level of the wine. + Sweetness: ( S ), where ( S ) represents integers indicating the range of sweetness levels. Example: 1-2 + Acidity_keyword: The exact keywords in the user's query describing the acidity level of the wine. + Acidity: ( A ), where ( A ) represents integers indicating the range of acidity level. Example: 3-5 + Tannin_keyword: The exact keywords in the user's query describing the tannin level of the wine. + Tannin: ( T ), where ( T ) represents integers indicating the range of tannin level. Example: 1-3 + Intensity_keyword: The exact keywords in the user's query describing the intensity level of the wine. + Intensity: ( I ), where ( I ) represents integers indicating the range of intensity level. Example: 2-4 + - You should only respond in the form (JSON) as described below: - { - "sweetness_keyword": ..., - "sweetness": ..., - "acidity_keyword": ..., - "acidity": ..., - "tannin_keyword": ..., - "tannin": ..., - "intensity_keyword": ..., - "intensity": ... - } + + Sweetness_keyword: ... + Sweetness: ... + Acidity_keyword: ... + Acidity: ... + Tannin_keyword: ... + Tannin: ... + Intensity_keyword: ... + Intensity: ... + - Here are some examples: - User's query: I want a wine with a medium-bodied, low acidity, medium tannin. - { - "sweetness_keyword": "NA", - "sweetness": "NA", - "acidity_keyword": "low acidity", - "acidity": "1-2", - "tannin_keyword": "medium tannin", - "tannin": "3-4", - "intensity_keyword": "medium-bodied", - "intensity": "3-4" - } - - - User's query: German red wine, under 100, pairs with spicy food - { - "sweetness_keyword": "NA", - "sweetness": "NA", - "acidity_keyword": "NA", - "acidity": "NA", - "tannin_keyword": "NA", - "tannin": "NA", - "intensity_keyword": "NA", - "intensity": "NA" - } + + User's query: I want a wine with a medium-bodied, low acidity, medium tannin. + Sweetness_keyword: NA + Sweetness: NA + Acidity_keyword: low acidity + Acidity: 1-2 + Tannin_keyword: medium tannin + Tannin: 3-4 + Intensity_keyword: medium-bodied + Intensity: 3-4 + User's query: German red wine, under 100, pairs with spicy food + Sweetness_keyword: NA + Sweetness: NA + Acidity_keyword: NA + Acidity: NA + Tannin_keyword: NA + Tannin: NA + Intensity_keyword: NA + Intensity: NA + Let's begin! """ - + header = ["Sweetness_keyword:", "Sweetness:", "Acidity_keyword:", "Acidity:", "Tannin_keyword:", "Tannin:", "Intensity_keyword:", "Intensity:"] + dictkey = ["sweetness_keyword", "sweetness", "acidity_keyword", "acidity", "tannin_keyword", "tannin", "intensity_keyword", "intensity"] errornote = "" - for attempt in 1:5 + for attempt in 1:10 usermsg = """ $conversiontable @@ -622,14 +634,22 @@ function extractWineAttributes_2(a::T1, input::T2)::String where {T1<:agent, T2< ] # put in model format - prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct") - prompt *= - """ - <|start_header_id|>assistant<|end_header_id|> - """ + prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen") response = a.func[:text2textInstructLLM](prompt) - responsedict = copy(JSON3.read(response)) + + # check whether response has all header + detected_kw = GeneralUtils.detect_keyword(header, response) + if sum(values(detected_kw)) < length(header) + errornote = "\nYiemAgent extractWineAttributes_2() response does not have all header" + continue + elseif sum(values(detected_kw)) > length(header) + errornote = "\nYiemAgent extractWineAttributes_2() response has duplicated header" + continue + end + + responsedict = GeneralUtils.textToDict(response, header; + dictKey=dictkey, symbolkey=true) # check whether each describing keyword is in the input to prevent halucination for i in ["sweetness", "acidity", "tannin", "intensity"] @@ -753,8 +773,11 @@ function paraphrase(text2textInstructLLM::Function, text::String) response = replace(response, '$' => "USD") response = replace(response, '`' => "") response = GeneralUtils.remove_french_accents(response) - responsedict = GeneralUtils.textToDict(response, ["Paraphrase"], - rightmarker=":", symbolkey=true, lowercasekey=true) + + header = ["Paraphrase:"] + dictkey = ["paraphrase"] + responsedict = GeneralUtils.textToDict(response, header; + dictKey=dictkey, symbolkey=true) for i ∈ [:paraphrase] if length(JSON3.write(responsedict[i])) == 0 diff --git a/test/config.json b/test/config.json index 7bf19e8..74c7a75 100644 --- a/test/config.json +++ b/test/config.json @@ -31,26 +31,46 @@ "description": "organization name" }, "externalservice": { - "text2textinstruct": { - "mqtttopic": "/loadbalancer/requestingservice", - "description": "text to text service with instruct LLM", - "llminfo": { - "name": "llama3instruct" - } - }, - "virtualWineCustomer_1": { - "mqtttopic": "/virtualenvironment/winecustomer", - "description": "text to text service with instruct LLM that act as wine customer", - "llminfo": { - "name": "llama3instruct" - } - }, - "text2textchat": { - "mqtttopic": "/loadbalancer/requestingservice", - "description": "text to text service with instruct LLM", - "llminfo": { - "name": "llama3instruct" - } - } + "loadbalancer": { + "mqtttopic": "/loadbalancer/requestingservice", + "description": "text to text service with instruct LLM" + }, + "text2textinstruct": { + "mqtttopic": "/loadbalancer/requestingservice", + "description": "text to text service with instruct LLM", + "llminfo": { + "name": "llama3instruct" + } + }, + "virtualWineCustomer_1": { + "mqtttopic": "/virtualenvironment/winecustomer", + "description": "text to text service with instruct LLM that act as wine customer", + "llminfo": { + "name": "llama3instruct" + } + }, + "text2textchat": { + "mqtttopic": "/loadbalancer/requestingservice", + "description": "text to text service with instruct LLM", + "llminfo": { + "name": "llama3instruct" + } + }, + "wineDB" : { + "description": "A wine database connection info for LibPQ client", + "host": "192.168.88.12", + "port": 10201, + "dbname": "wineDB", + "user": "yiemtechnologies", + "password": "yiemtechnologies@Postgres_0.0" + }, + "SQLVectorDB" : { + "description": "A wine database connection info for LibPQ client", + "host": "192.168.88.12", + "port": 10203, + "dbname": "SQLVectorDB", + "user": "yiemtechnologies", + "password": "yiemtechnologies@Postgres_0.0" + } } } \ No newline at end of file diff --git a/test/etc.jl b/test/etc.jl deleted file mode 100644 index bae1520..0000000 --- a/test/etc.jl +++ /dev/null @@ -1,9 +0,0 @@ -using GeneralUtils - -response = "trajectory_evaluation:\nThe trajectory is correct so far. The thought accurately reflects the user's question, and the action taken is a valid attempt to retrieve data from the database that matches the specified criteria.\n\nanswer_evaluation:\nThe observation provides information about two red wines from Bordeaux rive droite in France, which partially answers the question. However, it does not provide a complete answer as it only lists the wine names and characteristics, but does not explicitly state whether there are any other wines that match the criteria.\n\naccepted_as_answer: No\n\nscore: 6\nThe trajectory is mostly correct, but the observation does not fully address the question.\n\nsuggestion: Consider adding more filters or parameters to the database query to retrieve a complete list of wines that match the specified criteria." - -responsedict = GeneralUtils.textToDict(response, - ["trajectory_evaluation", "answer_evaluation", "accepted_as_answer", "score", "suggestion"], - rightmarker=":", symbolkey=true) - - diff --git a/test/runtests.jl b/test/runtests.jl new file mode 100644 index 0000000..e69de29 diff --git a/test/runtest.jl b/test/test_1.jl similarity index 72% rename from test/runtest.jl rename to test/test_1.jl index 524bfe6..aaef67f 100644 --- a/test/runtest.jl +++ b/test/test_1.jl @@ -1,272 +1,292 @@ -using Revise -using JSON, JSON3, Dates, UUIDs, PrettyPrinting, LibPQ, Base64, DataFrames -using YiemAgent, GeneralUtils -using Base.Threads - -# ---------------------------------------------- 100 --------------------------------------------- # - - - -# load config -config = JSON3.read("./test/config.json") -# config = copy(JSON3.read("../mountvolume/config.json")) - - -function executeSQL(sql::T) where {T<:AbstractString} - DBconnection = LibPQ.Connection("host=192.168.88.12 port=10201 dbname=wineDB user=yiemtechnologies password=yiemtechnologies@Postgres_0.0") - result = LibPQ.execute(DBconnection, sql) - close(DBconnection) - return result -end - -function executeSQLVectorDB(sql) - DBconnection = LibPQ.Connection("host=192.168.88.12 port=10203 dbname=SQLVectorDB user=yiemtechnologies password=yiemtechnologies@Postgres_0.0") - result = LibPQ.execute(DBconnection, sql) - close(DBconnection) - return result -end - -function text2textInstructLLM(prompt::String) - msgMeta = GeneralUtils.generate_msgMeta( - config[:externalservice][:text2textinstruct][:mqtttopic]; - msgPurpose="inference", - senderName="yiemagent", - senderId=string(uuid4()), - receiverName="text2textinstruct", - mqttBrokerAddress=config[:mqttServerInfo][:broker], - mqttBrokerPort=config[:mqttServerInfo][:port], - ) - - outgoingMsg = Dict( - :msgMeta => msgMeta, - :payload => Dict( - :text => prompt, - :kwargs => Dict( - :num_ctx => 16384, - :temperature => 0.2, - ) - ) - ) - - _response = GeneralUtils.sendReceiveMqttMsg(outgoingMsg; timeout=6000) - response = _response[:response][:text] - - return response -end - -# get text embedding from a LLM service -function getEmbedding(text::T) where {T<:AbstractString} - msgMeta = GeneralUtils.generate_msgMeta( - config[:externalservice][:text2textinstruct][:mqtttopic]; - msgPurpose="embedding", - senderName="yiemagent", - senderId=string(uuid4()), - receiverName="text2textinstruct", - mqttBrokerAddress=config[:mqttServerInfo][:broker], - mqttBrokerPort=config[:mqttServerInfo][:port], - ) - - outgoingMsg = Dict( - :msgMeta => msgMeta, - :payload => Dict( - :text => [text] # must be a vector of string - ) - ) - response = GeneralUtils.sendReceiveMqttMsg(outgoingMsg; timeout=6000) - embedding = response[:response][:embeddings] - return embedding -end - -function findSimilarTextFromVectorDB(text::T1, tablename::T2, embeddingColumnName::T3, - vectorDB::Function; limit::Integer=1 - )::DataFrame where {T1<:AbstractString, T2<:AbstractString, T3<:AbstractString} - - # get embedding from LLM service - embedding = getEmbedding(text)[1] - - # check whether there is close enough vector already store in vectorDB. if no, add, else skip - sql = """ - SELECT *, $embeddingColumnName <-> '$embedding' as distance - FROM $tablename - ORDER BY distance LIMIT $limit; - """ - response = vectorDB(sql) - df = DataFrame(response) - return df -end - - -function similarSQLVectorDB(query; maxdistance::Integer=100) - tablename = "sqlllm_decision_repository" - # get embedding of the query - df = findSimilarTextFromVectorDB(query, tablename, - "function_input_embedding", executeSQLVectorDB) - row, col = size(df) - distance = row == 0 ? Inf : df[1, :distance] - if row != 0 && distance < maxdistance - # if there is usable SQL, return it. - output_b64 = df[1, :function_output_base64] # pick the closest match - output_str = String(base64decode(output_b64)) - rowid = df[1, :id] - println("\n~~~ found similar sql. row id $rowid, distance $distance ", @__FILE__, " ", @__LINE__) - return (dict=output_str, distance=distance) - else - println("\n~~~ similar sql not found, max distance $maxdistance ", @__FILE__, " ", @__LINE__) - return (dict=nothing, distance=nothing) - end -end - - -function insertSQLVectorDB(query::T1, SQL::T2; maxdistance::Integer=1) where {T1<:AbstractString, T2<:AbstractString} - tablename = "sqlllm_decision_repository" - # get embedding of the query - # query = state[:thoughtHistory][:question] - df = findSimilarTextFromVectorDB(query, tablename, - "function_input_embedding", executeSQLVectorDB) - row, col = size(df) - distance = row == 0 ? Inf : df[1, :distance] - if row == 0 || distance > maxdistance # no close enough SQL stored in the database - query_embedding = getEmbedding(query)[1] - query = replace(query, "'" => "") - sql_base64 = base64encode(SQL) - sql_ = replace(SQL, "'" => "") - - sql = """ - INSERT INTO $tablename (function_input, function_output, function_output_base64, function_input_embedding) VALUES ('$query', '$sql_', '$sql_base64', '$query_embedding'); - """ - println("\n~~~ added new decision to vectorDB ", @__FILE__, " ", @__LINE__) - println(sql) - _ = executeSQLVectorDB(sql) - end -end - - -function similarSommelierDecision(recentevents::T1; maxdistance::Integer=5 - )::Union{AbstractDict, Nothing} where {T1<:AbstractString} - tablename = "sommelier_decision_repository" - # find similar - println("\n~~~ search vectorDB for this: $recentevents ", @__FILE__, " ", @__LINE__) - df = findSimilarTextFromVectorDB(recentevents, tablename, - "function_input_embedding", executeSQLVectorDB) - row, col = size(df) - distance = row == 0 ? Inf : df[1, :distance] - if row != 0 && distance < maxdistance - # if there is usable decision, return it. - rowid = df[1, :id] - println("\n~~~ found similar decision. row id $rowid, distance $distance ", @__FILE__, " ", @__LINE__) - output_b64 = df[1, :function_output_base64] # pick the closest match - _output_str = String(base64decode(output_b64)) - output = copy(JSON3.read(_output_str)) - return output - else - println("\n~~~ similar decision not found, max distance $maxdistance ", @__FILE__, " ", @__LINE__) - return nothing - end -end - - -function insertSommelierDecision(recentevents::T1, decision::T2; maxdistance::Integer=5 - ) where {T1<:AbstractString, T2<:AbstractDict} - tablename = "sommelier_decision_repository" - # find similar - df = findSimilarTextFromVectorDB(recentevents, tablename, - "function_input_embedding", executeSQLVectorDB) - row, col = size(df) - distance = row == 0 ? Inf : df[1, :distance] - if row == 0 || distance > maxdistance # no close enough SQL stored in the database - recentevents_embedding = a.func[:getEmbedding](recentevents)[1] - recentevents = replace(recentevents, "'" => "") - decision_json = JSON3.write(decision) - decision_base64 = base64encode(decision_json) - decision = replace(decision_json, "'" => "") - - sql = """ - INSERT INTO $tablename (function_input, function_output, function_output_base64, function_input_embedding) VALUES ('$recentevents', '$decision', '$decision_base64', '$recentevents_embedding'); - """ - println("\n~~~ added new decision to vectorDB ", @__FILE__, " ", @__LINE__) - println(sql) - _ = executeSQLVectorDB(sql) - else - println("~~~ similar decision previously cached, distance $distance ", @__FILE__, " ", @__LINE__) - end -end - - -sessionId = "12345" - -externalFunction = ( - getEmbedding=getEmbedding, - text2textInstructLLM=text2textInstructLLM, - executeSQL=executeSQL, - similarSQLVectorDB=similarSQLVectorDB, - insertSQLVectorDB=insertSQLVectorDB, - similarSommelierDecision=similarSommelierDecision, - insertSommelierDecision=insertSommelierDecision, - ) - - - -a = YiemAgent.sommelier( - externalFunction; - name="Ton", - id=sessionId, # agent instance id - retailername="Yiem", -) - -while true - println("your respond: ") - user_answer = readline() - response = YiemAgent.conversation(a, Dict(:text=> user_answer)) - println("\n$response") -end - - -# response = YiemAgent.conversation(a, Dict(:text=> "I want to get a French red wine under 100.")) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - +using Revise +using JSON, JSON3, Dates, UUIDs, PrettyPrinting, LibPQ, Base64, DataFrames +using YiemAgent, GeneralUtils +using Base.Threads + +# ---------------------------------------------- 100 --------------------------------------------- # + + + +# load config +config = JSON3.read("/appfolder/app/dev/YiemAgent/test/config.json") +# config = copy(JSON3.read("../mountvolume/config.json")) + + +function executeSQL(sql::T) where {T<:AbstractString} + host = config[:externalservice][:wineDB][:host] + port = config[:externalservice][:wineDB][:port] + dbname = config[:externalservice][:wineDB][:dbname] + user = config[:externalservice][:wineDB][:user] + password = config[:externalservice][:wineDB][:password] + DBconnection = LibPQ.Connection("host=$host port=$port dbname=$dbname user=$user password=$password") + result = LibPQ.execute(DBconnection, sql) + close(DBconnection) + return result +end + +function executeSQLVectorDB(sql) + host = config[:externalservice][:SQLVectorDB][:host] + port = config[:externalservice][:SQLVectorDB][:port] + dbname = config[:externalservice][:SQLVectorDB][:dbname] + user = config[:externalservice][:SQLVectorDB][:user] + password = config[:externalservice][:SQLVectorDB][:password] + DBconnection = LibPQ.Connection("host=$host port=$port dbname=$dbname user=$user password=$password") + result = LibPQ.execute(DBconnection, sql) + close(DBconnection) + return result +end + +function text2textInstructLLM(prompt::String; maxattempt=3) + msgMeta = GeneralUtils.generate_msgMeta( + config[:externalservice][:loadbalancer][:mqtttopic]; + msgPurpose="inference", + senderName="yiemagent", + senderId=sessionId, + receiverName="text2textinstruct_small", + mqttBrokerAddress=config[:mqttServerInfo][:broker], + mqttBrokerPort=config[:mqttServerInfo][:port], + ) + + outgoingMsg = Dict( + :msgMeta => msgMeta, + :payload => Dict( + :text => prompt, + :kwargs => Dict( + :num_ctx => 16384, + :temperature => 0.2, + ) + ) + ) + + response = nothing + for attempts in 1:maxattempt + _response = GeneralUtils.sendReceiveMqttMsg(outgoingMsg; timeout=300, maxattempt=maxattempt) + payload = _response[:response] + if _response[:success] && payload[:text] !== nothing + response = _response[:response][:text] + break + else + println("\n attempt $attempts/$maxattempt failed ", @__FILE__, ":", @__LINE__, " $(Dates.now())") + pprintln(outgoingMsg) + println(" attempt $attempts/$maxattempt failed ", @__FILE__, ":", @__LINE__, " $(Dates.now())\n") + sleep(3) + end + end + + return response +end + +# get text embedding from a LLM service +function getEmbedding(text::T) where {T<:AbstractString} + msgMeta = GeneralUtils.generate_msgMeta( + config[:externalservice][:loadbalancer][:mqtttopic]; + msgPurpose="embedding", + senderName="yiemagent", + senderId=sessionId, + receiverName="text2textinstruct_small", + mqttBrokerAddress=config[:mqttServerInfo][:broker], + mqttBrokerPort=config[:mqttServerInfo][:port], + ) + + outgoingMsg = Dict( + :msgMeta => msgMeta, + :payload => Dict( + :text => [text] # must be a vector of string + ) + ) + response = GeneralUtils.sendReceiveMqttMsg(outgoingMsg; timeout=120) + embedding = response[:response][:embeddings] + return embedding +end + +function findSimilarTextFromVectorDB(text::T1, tablename::T2, embeddingColumnName::T3, + vectorDB::Function; limit::Integer=1 + )::DataFrame where {T1<:AbstractString, T2<:AbstractString, T3<:AbstractString} + # get embedding from LLM service + embedding = getEmbedding(text)[1] + # check whether there is close enough vector already store in vectorDB. if no, add, else skip + sql = """ + SELECT *, $embeddingColumnName <-> '$embedding' as distance + FROM $tablename + ORDER BY distance LIMIT $limit; + """ + response = vectorDB(sql) + df = DataFrame(response) + return df +end + +function similarSQLVectorDB(query; maxdistance::Integer=100) + tablename = "sqlllm_decision_repository" + # get embedding of the query + df = findSimilarTextFromVectorDB(query, tablename, + "function_input_embedding", executeSQLVectorDB) + # println(df[1, [:id, :function_output]]) + row, col = size(df) + distance = row == 0 ? Inf : df[1, :distance] + # distance = 100 # CHANGE this is for testing only + if row != 0 && distance < maxdistance + # if there is usable SQL, return it. + output_b64 = df[1, :function_output_base64] # pick the closest match + output_str = String(base64decode(output_b64)) + rowid = df[1, :id] + println("\n~~~ found similar sql. row id $rowid, distance $distance ", @__FILE__, ":", @__LINE__, " $(Dates.now())") + return (dict=output_str, distance=distance) + else + println("\n~~~ similar sql not found, max distance $maxdistance ", @__FILE__, ":", @__LINE__, " $(Dates.now())") + return (dict=nothing, distance=nothing) + end +end + +function insertSQLVectorDB(query::T1, SQL::T2; maxdistance::Integer=3) where {T1<:AbstractString, T2<:AbstractString} + tablename = "sqlllm_decision_repository" + # get embedding of the query + # query = state[:thoughtHistory][:question] + df = findSimilarTextFromVectorDB(query, tablename, + "function_input_embedding", executeSQLVectorDB) + row, col = size(df) + distance = row == 0 ? Inf : df[1, :distance] + if row == 0 || distance > maxdistance # no close enough SQL stored in the database + query_embedding = getEmbedding(query)[1] + query = replace(query, "'" => "") + sql_base64 = base64encode(SQL) + sql_ = replace(SQL, "'" => "") + + sql = """ + INSERT INTO $tablename (function_input, function_output, function_output_base64, function_input_embedding) VALUES ('$query', '$sql_', '$sql_base64', '$query_embedding'); + """ + # println("\n~~~ added new decision to vectorDB ", @__FILE__, ":", @__LINE__, " $(Dates.now())") + # println(sql) + _ = executeSQLVectorDB(sql) + end +end + + +function similarSommelierDecision(recentevents::T1; maxdistance::Integer=5 + )::Union{AbstractDict, Nothing} where {T1<:AbstractString} + tablename = "sommelier_decision_repository" + # find similar + println("\n~~~ search vectorDB for this: $recentevents ", @__FILE__, " ", @__LINE__) + df = findSimilarTextFromVectorDB(recentevents, tablename, + "function_input_embedding", executeSQLVectorDB) + row, col = size(df) + distance = row == 0 ? Inf : df[1, :distance] + if row != 0 && distance < maxdistance + # if there is usable decision, return it. + rowid = df[1, :id] + println("\n~~~ found similar decision. row id $rowid, distance $distance ", @__FILE__, " ", @__LINE__) + output_b64 = df[1, :function_output_base64] # pick the closest match + _output_str = String(base64decode(output_b64)) + output = copy(JSON3.read(_output_str)) + return output + else + println("\n~~~ similar decision not found, max distance $maxdistance ", @__FILE__, " ", @__LINE__) + return nothing + end +end + + +function insertSommelierDecision(recentevents::T1, decision::T2; maxdistance::Integer=5 + ) where {T1<:AbstractString, T2<:AbstractDict} + tablename = "sommelier_decision_repository" + # find similar + df = findSimilarTextFromVectorDB(recentevents, tablename, + "function_input_embedding", executeSQLVectorDB) + row, col = size(df) + distance = row == 0 ? Inf : df[1, :distance] + if row == 0 || distance > maxdistance # no close enough SQL stored in the database + recentevents_embedding = a.func[:getEmbedding](recentevents)[1] + recentevents = replace(recentevents, "'" => "") + decision_json = JSON3.write(decision) + decision_base64 = base64encode(decision_json) + decision = replace(decision_json, "'" => "") + + sql = """ + INSERT INTO $tablename (function_input, function_output, function_output_base64, function_input_embedding) VALUES ('$recentevents', '$decision', '$decision_base64', '$recentevents_embedding'); + """ + println("\n~~~ added new decision to vectorDB ", @__FILE__, " ", @__LINE__) + println(sql) + _ = executeSQLVectorDB(sql) + else + println("~~~ similar decision previously cached, distance $distance ", @__FILE__, " ", @__LINE__) + end +end + + +sessionId = "12345" + +externalFunction = ( + getEmbedding=getEmbedding, + text2textInstructLLM=text2textInstructLLM, + executeSQL=executeSQL, + similarSQLVectorDB=similarSQLVectorDB, + insertSQLVectorDB=insertSQLVectorDB, + similarSommelierDecision=similarSommelierDecision, + insertSommelierDecision=insertSommelierDecision, + ) + + + +a = YiemAgent.sommelier( + externalFunction; + name="Ton", + id=sessionId, # agent instance id + retailername="Yiem", +) + +while true + println("your respond: ") + user_answer = readline() + response = YiemAgent.conversation(a, Dict(:text=> user_answer)) + println("\n$response") +end + + +# response = YiemAgent.conversation(a, Dict(:text=> "I want to get a French red wine under 100.")) + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +