Starting to work again
This commit is contained in:
parent
7f24d8870c
commit
2a3dc56897
738
frontend/package-lock.json
generated
738
frontend/package-lock.json
generated
File diff suppressed because it is too large
Load Diff
@ -18,6 +18,7 @@
|
||||
"@types/node": "^16.18.126",
|
||||
"@types/react": "^19.0.12",
|
||||
"@types/react-dom": "^19.0.4",
|
||||
"@uiw/react-json-view": "^2.0.0-alpha.31",
|
||||
"mui-markdown": "^1.2.6",
|
||||
"react": "^19.0.0",
|
||||
"react-dom": "^19.0.0",
|
||||
|
||||
@ -26,8 +26,8 @@ interface ConversationHandle {
|
||||
|
||||
interface BackstoryMessage {
|
||||
prompt: string;
|
||||
preamble: string;
|
||||
content: string;
|
||||
preamble: {};
|
||||
full_content: string;
|
||||
response: string;
|
||||
metadata: {
|
||||
rag: { documents: [] };
|
||||
@ -206,8 +206,8 @@ const Conversation = forwardRef<ConversationHandle, ConversationProps>(({
|
||||
}, {
|
||||
role: 'assistant',
|
||||
prompt: message.prompt || "",
|
||||
preamble: message.preamble || "",
|
||||
full_content: message.content || "",
|
||||
preamble: message.preamble || {},
|
||||
full_content: message.full_content || "",
|
||||
content: message.response || "",
|
||||
metadata: message.metadata,
|
||||
actions: message.actions,
|
||||
@ -402,53 +402,59 @@ const Conversation = forwardRef<ConversationHandle, ConversationProps>(({
|
||||
|
||||
try {
|
||||
const update = JSON.parse(line);
|
||||
console.log('Parsed update:', update.response);
|
||||
|
||||
// Force an immediate state update based on the message type
|
||||
if (update.status === 'processing') {
|
||||
// Update processing message with immediate re-render
|
||||
setProcessingMessage({ role: 'status', content: update.message });
|
||||
// Add a small delay to ensure React has time to update the UI
|
||||
await new Promise(resolve => setTimeout(resolve, 0));
|
||||
} else if (update.status === 'done') {
|
||||
// Replace processing message with final result
|
||||
if (onResponse) {
|
||||
update.message = onResponse(update.message);
|
||||
}
|
||||
setProcessingMessage(undefined);
|
||||
const backstoryMessage: BackstoryMessage = update.message;
|
||||
setConversation([
|
||||
...conversationRef.current, {
|
||||
role: 'user',
|
||||
content: backstoryMessage.prompt || "",
|
||||
}, {
|
||||
role: 'assistant',
|
||||
prompt: backstoryMessage.prompt || "",
|
||||
preamble: backstoryMessage.preamble || "",
|
||||
full_content: backstoryMessage.content || "",
|
||||
content: backstoryMessage.response || "",
|
||||
metadata: backstoryMessage.metadata,
|
||||
actions: backstoryMessage.actions,
|
||||
}] as MessageList);
|
||||
// Add a small delay to ensure React has time to update the UI
|
||||
await new Promise(resolve => setTimeout(resolve, 0));
|
||||
|
||||
const metadata = update.message.metadata;
|
||||
if (metadata) {
|
||||
const evalTPS = metadata.eval_count * 10 ** 9 / metadata.eval_duration;
|
||||
const promptTPS = metadata.prompt_eval_count * 10 ** 9 / metadata.prompt_eval_duration;
|
||||
setLastEvalTPS(evalTPS ? evalTPS : 35);
|
||||
setLastPromptTPS(promptTPS ? promptTPS : 35);
|
||||
updateContextStatus();
|
||||
}
|
||||
} else if (update.status === 'error') {
|
||||
// Show error
|
||||
setProcessingMessage({ role: 'error', content: update.message });
|
||||
setTimeout(() => {
|
||||
switch (update.status) {
|
||||
case 'processing':
|
||||
case 'thinking':
|
||||
// Force an immediate state update based on the message type
|
||||
// Update processing message with immediate re-render
|
||||
setProcessingMessage({ role: 'status', content: update.response });
|
||||
// Add a small delay to ensure React has time to update the UI
|
||||
await new Promise(resolve => setTimeout(resolve, 0));
|
||||
break;
|
||||
case 'done':
|
||||
// Replace processing message with final result
|
||||
if (onResponse) {
|
||||
update.message = onResponse(update.response);
|
||||
}
|
||||
setProcessingMessage(undefined);
|
||||
}, 5000);
|
||||
const backstoryMessage: BackstoryMessage = update.response;
|
||||
setConversation([
|
||||
...conversationRef.current, {
|
||||
role: 'user',
|
||||
content: backstoryMessage.prompt || "",
|
||||
}, {
|
||||
role: 'assistant',
|
||||
prompt: backstoryMessage.prompt || "",
|
||||
preamble: backstoryMessage.preamble || {},
|
||||
full_content: backstoryMessage.full_content || "",
|
||||
content: backstoryMessage.response || "",
|
||||
metadata: backstoryMessage.metadata,
|
||||
actions: backstoryMessage.actions,
|
||||
}] as MessageList);
|
||||
// Add a small delay to ensure React has time to update the UI
|
||||
await new Promise(resolve => setTimeout(resolve, 0));
|
||||
|
||||
// Add a small delay to ensure React has time to update the UI
|
||||
await new Promise(resolve => setTimeout(resolve, 0));
|
||||
const metadata = update.metadata;
|
||||
if (metadata) {
|
||||
const evalTPS = metadata.eval_count * 10 ** 9 / metadata.eval_duration;
|
||||
const promptTPS = metadata.prompt_eval_count * 10 ** 9 / metadata.prompt_eval_duration;
|
||||
setLastEvalTPS(evalTPS ? evalTPS : 35);
|
||||
setLastPromptTPS(promptTPS ? promptTPS : 35);
|
||||
updateContextStatus();
|
||||
}
|
||||
break;
|
||||
case 'error':
|
||||
// Show error
|
||||
setProcessingMessage({ role: 'error', content: update.response });
|
||||
setTimeout(() => {
|
||||
setProcessingMessage(undefined);
|
||||
}, 5000);
|
||||
|
||||
// Add a small delay to ensure React has time to update the UI
|
||||
await new Promise(resolve => setTimeout(resolve, 0));
|
||||
break;
|
||||
}
|
||||
} catch (e) {
|
||||
setSnack("Error processing query", "error")
|
||||
@ -462,25 +468,42 @@ const Conversation = forwardRef<ConversationHandle, ConversationProps>(({
|
||||
try {
|
||||
const update = JSON.parse(buffer);
|
||||
|
||||
if (update.status === 'done') {
|
||||
if (onResponse) {
|
||||
update.message = onResponse(update.message);
|
||||
}
|
||||
setProcessingMessage(undefined);
|
||||
const backstoryMessage: BackstoryMessage = update.message;
|
||||
setConversation([
|
||||
...conversationRef.current, {
|
||||
role: 'user',
|
||||
content: backstoryMessage.prompt || "",
|
||||
}, {
|
||||
role: 'assistant',
|
||||
prompt: backstoryMessage.prompt || "",
|
||||
preamble: backstoryMessage.preamble || "",
|
||||
full_content: backstoryMessage.content || "",
|
||||
content: backstoryMessage.response || "",
|
||||
metadata: backstoryMessage.metadata,
|
||||
actions: backstoryMessage.actions,
|
||||
}] as MessageList);
|
||||
switch (update.status) {
|
||||
case 'processing':
|
||||
case 'thinking':
|
||||
// Force an immediate state update based on the message type
|
||||
// Update processing message with immediate re-render
|
||||
setProcessingMessage({ role: 'status', content: update.response });
|
||||
// Add a small delay to ensure React has time to update the UI
|
||||
await new Promise(resolve => setTimeout(resolve, 0));
|
||||
break;
|
||||
case 'error':
|
||||
// Show error
|
||||
setProcessingMessage({ role: 'error', content: update.response });
|
||||
setTimeout(() => {
|
||||
setProcessingMessage(undefined);
|
||||
}, 5000);
|
||||
break;
|
||||
case 'done':
|
||||
if (onResponse) {
|
||||
update.message = onResponse(update.message);
|
||||
}
|
||||
setProcessingMessage(undefined);
|
||||
const backstoryMessage: BackstoryMessage = update.message;
|
||||
setConversation([
|
||||
...conversationRef.current, {
|
||||
role: 'user',
|
||||
content: backstoryMessage.prompt || "",
|
||||
}, {
|
||||
role: 'assistant',
|
||||
prompt: backstoryMessage.prompt || "",
|
||||
preamble: backstoryMessage.preamble || {},
|
||||
full_content: backstoryMessage.full_content || "",
|
||||
content: backstoryMessage.response || "",
|
||||
metadata: backstoryMessage.metadata,
|
||||
actions: backstoryMessage.actions,
|
||||
}] as MessageList);
|
||||
break;
|
||||
}
|
||||
} catch (e) {
|
||||
setSnack("Error processing query", "error")
|
||||
|
||||
@ -32,6 +32,8 @@ type MessageRoles = 'info' | 'user' | 'assistant' | 'system' | 'status' | 'error
|
||||
type MessageData = {
|
||||
role: MessageRoles,
|
||||
content: string,
|
||||
full_content?: string,
|
||||
|
||||
disableCopy?: boolean,
|
||||
user?: string,
|
||||
title?: string,
|
||||
@ -48,7 +50,6 @@ interface MessageMetaData {
|
||||
vector_embedding: number[];
|
||||
},
|
||||
origin: string,
|
||||
full_query?: string,
|
||||
rag: any,
|
||||
tools: any[],
|
||||
eval_count: number,
|
||||
@ -87,7 +88,6 @@ interface MessageMetaProps {
|
||||
const MessageMeta = (props: MessageMetaProps) => {
|
||||
const {
|
||||
/* MessageData */
|
||||
full_query,
|
||||
rag,
|
||||
tools,
|
||||
eval_count,
|
||||
@ -95,7 +95,7 @@ const MessageMeta = (props: MessageMetaProps) => {
|
||||
prompt_eval_count,
|
||||
prompt_eval_duration,
|
||||
} = props.metadata || {};
|
||||
const messageProps = props.messageProps;
|
||||
const message = props.messageProps.message;
|
||||
|
||||
return (<>
|
||||
<Box sx={{ fontSize: "0.8rem", mb: 1 }}>
|
||||
@ -137,7 +137,7 @@ const MessageMeta = (props: MessageMetaProps) => {
|
||||
</TableContainer>
|
||||
|
||||
{
|
||||
full_query !== undefined &&
|
||||
message.full_content !== undefined &&
|
||||
<Accordion>
|
||||
<AccordionSummary expandIcon={<ExpandMoreIcon />}>
|
||||
<Box sx={{ fontSize: "0.8rem" }}>
|
||||
@ -145,7 +145,7 @@ const MessageMeta = (props: MessageMetaProps) => {
|
||||
</Box>
|
||||
</AccordionSummary>
|
||||
<AccordionDetails>
|
||||
<pre style={{ "display": "block", "position": "relative" }}><CopyBubble content={full_query?.trim()} />{full_query?.trim()}</pre>
|
||||
<pre style={{ "display": "block", "position": "relative" }}><CopyBubble content={message.full_content?.trim()} />{message.full_content?.trim()}</pre>
|
||||
</AccordionDetails>
|
||||
</Accordion>
|
||||
}
|
||||
@ -182,8 +182,8 @@ const MessageMeta = (props: MessageMetaProps) => {
|
||||
</Accordion>
|
||||
}
|
||||
{
|
||||
rag?.name !== undefined && <>
|
||||
<Accordion>
|
||||
rag.map((rag: any) => (<>
|
||||
<Accordion key={rag.name}>
|
||||
<AccordionSummary expandIcon={<ExpandMoreIcon />}>
|
||||
<Box sx={{ fontSize: "0.8rem" }}>
|
||||
Top RAG {rag.ids.length} matches from '{rag.name}' collection against embedding vector of {rag.query_embedding.length} dimensions
|
||||
@ -212,7 +212,7 @@ const MessageMeta = (props: MessageMetaProps) => {
|
||||
</Box>
|
||||
</AccordionSummary>
|
||||
<AccordionDetails>
|
||||
<VectorVisualizer inline {...messageProps} {...props.metadata} rag={rag} />
|
||||
<VectorVisualizer inline {...message} {...props.metadata} rag={rag} />
|
||||
</AccordionDetails>
|
||||
</Accordion>
|
||||
<Accordion>
|
||||
@ -222,7 +222,7 @@ const MessageMeta = (props: MessageMetaProps) => {
|
||||
</Box>
|
||||
</AccordionSummary>
|
||||
<AccordionDetails>
|
||||
{Object.entries(props.messageProps.message)
|
||||
{Object.entries(message)
|
||||
.filter(([key, value]) => key !== undefined && value !== undefined)
|
||||
.map(([key, value]) => (typeof (value) !== "string" || value?.trim() !== "") &&
|
||||
<Accordion key={key}>
|
||||
@ -251,8 +251,7 @@ const MessageMeta = (props: MessageMetaProps) => {
|
||||
)}
|
||||
</AccordionDetails>
|
||||
</Accordion>
|
||||
|
||||
</>
|
||||
</>))
|
||||
}
|
||||
</>);
|
||||
};
|
||||
|
||||
@ -970,23 +970,6 @@ class WebServer:
|
||||
else:
|
||||
yield {"status": "complete", "message": "RAG processing complete"}
|
||||
|
||||
# agent_type: chat
|
||||
# * Q&A
|
||||
#
|
||||
# agent_type: job_description
|
||||
# * First message sets Job Description and generates Resume
|
||||
# * Has content (Job Description)
|
||||
# * Then Q&A of Job Description
|
||||
#
|
||||
# agent_type: resume
|
||||
# * First message sets Resume and generates Fact Check
|
||||
# * Has no content
|
||||
# * Then Q&A of Resume
|
||||
#
|
||||
# Fact Check:
|
||||
# * First message sets Fact Check and is Q&A
|
||||
# * Has content
|
||||
# * Then Q&A of Fact Check
|
||||
async def generate_response(self, context : Context, agent : Agent, content : str) -> AsyncGenerator[Message, None]:
|
||||
if not self.file_watcher:
|
||||
raise Exception("File watcher not initialized")
|
||||
@ -995,24 +978,20 @@ class WebServer:
|
||||
logger.info(f"generate_response: {agent_type}")
|
||||
if agent_type == "chat":
|
||||
message = Message(prompt=content)
|
||||
async for value in agent.prepare_message(message):
|
||||
async for message in agent.prepare_message(message):
|
||||
# logger.info(f"{agent_type}.prepare_message: {value.status} - {value.response}")
|
||||
if value.status != "done":
|
||||
yield value
|
||||
if value.status == "error":
|
||||
message.status = "error"
|
||||
message.response = value.response
|
||||
if message.status == "error":
|
||||
yield message
|
||||
return
|
||||
async for value in agent.process_message(self.llm, self.model, message):
|
||||
if message.status != "done":
|
||||
yield message
|
||||
async for message in agent.process_message(self.llm, self.model, message):
|
||||
# logger.info(f"{agent_type}.process_message: {value.status} - {value.response}")
|
||||
if value.status != "done":
|
||||
yield value
|
||||
if value.status == "error":
|
||||
message.status = "error"
|
||||
message.response = value.response
|
||||
if message.status == "error":
|
||||
yield message
|
||||
return
|
||||
if message.status != "done":
|
||||
yield message
|
||||
# async for value in agent.generate_llm_response(message):
|
||||
# logger.info(f"{agent_type}.generate_llm_response: {value.status} - {value.response}")
|
||||
# if value.status != "done":
|
||||
@ -1023,6 +1002,7 @@ class WebServer:
|
||||
# yield message
|
||||
# return
|
||||
logger.info("TODO: There is more to do...")
|
||||
yield message
|
||||
return
|
||||
|
||||
return
|
||||
|
||||
@ -30,15 +30,13 @@ class Chat(Agent, ABC):
|
||||
if not message.disable_rag:
|
||||
# Gather RAG results, yielding each result
|
||||
# as it becomes available
|
||||
for value in self.context.generate_rag_results(message):
|
||||
logging.info(f"RAG: {value.status} - {value.response}")
|
||||
if value.status != "done":
|
||||
yield value
|
||||
if value.status == "error":
|
||||
message.status = "error"
|
||||
message.response = value.response
|
||||
for message in self.context.generate_rag_results(message):
|
||||
logging.info(f"RAG: {message.status} - {message.response}")
|
||||
if message.status == "error":
|
||||
yield message
|
||||
return
|
||||
if message.status != "done":
|
||||
yield message
|
||||
|
||||
if "rag" in message.metadata and message.metadata["rag"]:
|
||||
for rag in message.metadata["rag"]:
|
||||
@ -106,11 +104,13 @@ class Chat(Agent, ABC):
|
||||
yield message
|
||||
if value.done:
|
||||
response = value
|
||||
message.status = "done"
|
||||
|
||||
if not response:
|
||||
message.status = "error"
|
||||
message.response = "No response from LLM."
|
||||
yield message
|
||||
self.context.processing = False
|
||||
return
|
||||
|
||||
message.metadata["eval_count"] += response["eval_count"]
|
||||
@ -120,6 +120,7 @@ class Chat(Agent, ABC):
|
||||
self.context_tokens = response["prompt_eval_count"] + response["eval_count"]
|
||||
|
||||
yield message
|
||||
self.context.processing = False
|
||||
return
|
||||
|
||||
tools_used = []
|
||||
@ -205,6 +206,7 @@ class Chat(Agent, ABC):
|
||||
message.full_content = ""
|
||||
for i, p in enumerate(message.preamble.keys()):
|
||||
message.full_content += '' if i == 0 else '\n\n' + f"<|{p}|>{message.preamble[p].strip()}\n"
|
||||
message.full_content += f"{message.prompt}"
|
||||
|
||||
# Estimate token length of new messages
|
||||
message.metadata["ctx_size"] = self.context.get_optimal_ctx_size(self.context_tokens, messages=message.full_content)
|
||||
@ -213,12 +215,14 @@ class Chat(Agent, ABC):
|
||||
message.status = "thinking"
|
||||
yield message
|
||||
|
||||
async for value in self.generate_llm_response(llm, model, message):
|
||||
logging.info(f"LLM: {value.status} - {value.response}")
|
||||
if value.status != "done":
|
||||
yield value
|
||||
if value.status == "error":
|
||||
async for message in self.generate_llm_response(llm, model, message):
|
||||
logging.info(f"LLM: {message.status} - {message.response}")
|
||||
if message.status == "error":
|
||||
return
|
||||
if message.status != "done":
|
||||
yield message
|
||||
yield message
|
||||
return
|
||||
|
||||
def get_and_reset_content_seed(self):
|
||||
tmp = self._content_seed
|
||||
|
||||
@ -119,8 +119,8 @@ class Context(BaseModel):
|
||||
yield message
|
||||
return
|
||||
except Exception as e:
|
||||
message.response = f"Error generating RAG results: {str(e)}"
|
||||
message.status = "error"
|
||||
message.response = f"Error generating RAG results: {str(e)}"
|
||||
logger.error(e)
|
||||
yield message
|
||||
return
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user