v4.6.9-alpha (#918)

Co-authored-by: Mufei <327958099@qq.com>
Co-authored-by: heheer <71265218+newfish-cmyk@users.noreply.github.com>
This commit is contained in:
Archer
2024-03-04 00:05:25 +08:00
committed by GitHub
parent f9f0b4bffd
commit 42a8184ea0
153 changed files with 4906 additions and 4307 deletions

View File

@@ -2,8 +2,10 @@ import { insertData2Dataset } from '@/service/core/dataset/data/controller';
import { MongoDatasetTraining } from '@fastgpt/service/core/dataset/training/schema';
import { TrainingModeEnum } from '@fastgpt/global/core/dataset/constants';
import { pushGenerateVectorUsage } from '@/service/support/wallet/usage/push';
import { checkInvalidChunkAndLock, checkTeamAiPointsAndLock } from './utils';
import { delay } from '@fastgpt/global/common/system/utils';
import { checkTeamAiPointsAndLock } from './utils';
import { checkInvalidChunkAndLock } from '@fastgpt/service/core/dataset/training/utils';
import { addMinutes } from 'date-fns';
import { addLog } from '@fastgpt/service/common/system/log';
const reduceQueue = () => {
global.vectorQueueLen = global.vectorQueueLen > 0 ? global.vectorQueueLen - 1 : 0;
@@ -13,7 +15,8 @@ const reduceQueue = () => {
/* 索引生成队列。每导入一次,就是一个单独的线程 */
export async function generateVector(): Promise<any> {
if (global.vectorQueueLen >= global.systemEnv.vectorMaxProcess) return;
const max = global.systemEnv?.vectorMaxProcess || 10;
if (global.vectorQueueLen >= max) return;
global.vectorQueueLen++;
const start = Date.now();
@@ -27,7 +30,7 @@ export async function generateVector(): Promise<any> {
try {
const data = await MongoDatasetTraining.findOneAndUpdate(
{
lockTime: { $lte: new Date(Date.now() - 1 * 60 * 1000) },
lockTime: { $lte: addMinutes(new Date(), -1) },
mode: TrainingModeEnum.chunk
},
{
@@ -68,7 +71,7 @@ export async function generateVector(): Promise<any> {
}
};
} catch (error) {
console.log(`Get Training Data error`, error);
addLog.error(`Get Training Data error`, error);
return {
error: true
};
@@ -77,11 +80,12 @@ export async function generateVector(): Promise<any> {
if (done || !data) {
if (reduceQueue()) {
console.log(`【index】Task done`);
addLog.info(`[Vector Queue] Done`);
}
return;
}
if (error) {
addLog.error(`[Vector Queue] Error`, { error });
reduceQueue();
return generateVector();
}
@@ -92,6 +96,8 @@ export async function generateVector(): Promise<any> {
return generateVector();
}
addLog.info(`[Vector Queue] Start`);
// create vector and insert
try {
// invalid data
@@ -103,7 +109,7 @@ export async function generateVector(): Promise<any> {
}
// insert to dataset
const { charsLength } = await insertData2Dataset({
const { tokens } = await insertData2Dataset({
teamId: data.teamId,
tmbId: data.tmbId,
datasetId: data.datasetId,
@@ -119,7 +125,7 @@ export async function generateVector(): Promise<any> {
pushGenerateVectorUsage({
teamId: data.teamId,
tmbId: data.tmbId,
charsLength,
tokens,
model: data.model,
billId: data.billId
});
@@ -129,7 +135,9 @@ export async function generateVector(): Promise<any> {
reduceQueue();
generateVector();
console.log(`embedding finished, time: ${Date.now() - start}ms`);
addLog.info(`[Vector Queue] Finish`, {
time: Date.now() - start
});
} catch (err: any) {
reduceQueue();