Trigger llm output retrieval on server (#39)

* Rename tables, add graphile workers, update types

* Add dev:worker command

* Update pnpm-lock.yaml

* Remove sentry config import from worker.ts

* Stop generating new cells in cell router get query

* Generate new cells for new scenarios, variants, and experiments

* Remove most error throwing from queryLLM.task.ts

* Remove promptVariantId and testScenarioId from ModelOutput

* Remove duplicate index from ModelOutput

* Move inputHash from cell to output

* Add TODO

* Add todo

* Show cost and time for each cell

* Always show output stats if there is output

* Trigger LLM outputs when scenario variables are updated

* Add newlines to ends of files

* Add another newline

* Cascade ModelOutput deletion

* Fix linting and prettier

* Return instead of throwing for non-pending cell

* Remove pnpm dev:worker from pnpm:dev

* Update pnpm-lock.yaml
This commit is contained in:
arcticfly
2023-07-14 16:38:46 -06:00
committed by GitHub
parent 032c07ec65
commit b98eb9b729
29 changed files with 1089 additions and 407 deletions

View File

@@ -0,0 +1,49 @@
-- Drop the foreign key constraints on the original ModelOutput
ALTER TABLE "ModelOutput" DROP CONSTRAINT "ModelOutput_promptVariantId_fkey";
ALTER TABLE "ModelOutput" DROP CONSTRAINT "ModelOutput_testScenarioId_fkey";
-- Rename the old table
ALTER TABLE "ModelOutput" RENAME TO "ScenarioVariantCell";
ALTER TABLE "ScenarioVariantCell" RENAME CONSTRAINT "ModelOutput_pkey" TO "ScenarioVariantCell_pkey";
ALTER INDEX "ModelOutput_inputHash_idx" RENAME TO "ScenarioVariantCell_inputHash_idx";
ALTER INDEX "ModelOutput_promptVariantId_testScenarioId_key" RENAME TO "ScenarioVariantCell_promptVariantId_testScenarioId_key";
-- Add the new fields to the renamed table
ALTER TABLE "ScenarioVariantCell" ADD COLUMN "retryTime" TIMESTAMP(3);
ALTER TABLE "ScenarioVariantCell" ADD COLUMN "streamingChannel" TEXT;
ALTER TABLE "ScenarioVariantCell" ALTER COLUMN "inputHash" DROP NOT NULL;
ALTER TABLE "ScenarioVariantCell" ALTER COLUMN "output" DROP NOT NULL,
ALTER COLUMN "statusCode" DROP NOT NULL,
ALTER COLUMN "timeToComplete" DROP NOT NULL;
-- Create the new table
CREATE TABLE "ModelOutput" (
"id" UUID NOT NULL,
"inputHash" TEXT NOT NULL,
"output" JSONB NOT NULL,
"timeToComplete" INTEGER NOT NULL DEFAULT 0,
"promptTokens" INTEGER,
"completionTokens" INTEGER,
"createdAt" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP,
"updatedAt" TIMESTAMP(3) NOT NULL,
"scenarioVariantCellId" UUID
);
-- Move inputHash index
DROP INDEX "ScenarioVariantCell_inputHash_idx";
CREATE INDEX "ModelOutput_inputHash_idx" ON "ModelOutput"("inputHash");
CREATE UNIQUE INDEX "ModelOutput_scenarioVariantCellId_key" ON "ModelOutput"("scenarioVariantCellId");
ALTER TABLE "ModelOutput" ADD CONSTRAINT "ModelOutput_scenarioVariantCellId_fkey" FOREIGN KEY ("scenarioVariantCellId") REFERENCES "ScenarioVariantCell"("id") ON DELETE CASCADE ON UPDATE CASCADE;
ALTER TABLE "ModelOutput" ALTER COLUMN "scenarioVariantCellId" SET NOT NULL,
ADD CONSTRAINT "ModelOutput_pkey" PRIMARY KEY ("id");
ALTER TABLE "ScenarioVariantCell" ADD CONSTRAINT "ScenarioVariantCell_promptVariantId_fkey" FOREIGN KEY ("promptVariantId") REFERENCES "PromptVariant"("id") ON DELETE CASCADE ON UPDATE CASCADE;
ALTER TABLE "ScenarioVariantCell" ADD CONSTRAINT "ScenarioVariantCell_testScenarioId_fkey" FOREIGN KEY ("testScenarioId") REFERENCES "TestScenario"("id") ON DELETE CASCADE ON UPDATE CASCADE;
-- CreateEnum
CREATE TYPE "CellRetrievalStatus" AS ENUM ('PENDING', 'IN_PROGRESS', 'COMPLETE', 'ERROR');
-- AlterTable
ALTER TABLE "ScenarioVariantCell" ADD COLUMN "retrievalStatus" "CellRetrievalStatus" NOT NULL DEFAULT 'COMPLETE';

View File

@@ -0,0 +1,2 @@
-- AlterTable
ALTER TABLE "PromptVariant" ADD COLUMN "model" TEXT NOT NULL DEFAULT 'gpt-3.5-turbo';

View File

@@ -41,7 +41,7 @@ model PromptVariant {
createdAt DateTime @default(now())
updatedAt DateTime @updatedAt
ModelOutput ModelOutput[]
scenarioVariantCells ScenarioVariantCell[]
EvaluationResult EvaluationResult[]
@@index([uiId])
@@ -61,7 +61,7 @@ model TestScenario {
createdAt DateTime @default(now())
updatedAt DateTime @updatedAt
ModelOutput ModelOutput[]
scenarioVariantCells ScenarioVariantCell[]
}
model TemplateVariable {
@@ -76,17 +76,28 @@ model TemplateVariable {
updatedAt DateTime @updatedAt
}
model ModelOutput {
enum CellRetrievalStatus {
PENDING
IN_PROGRESS
COMPLETE
ERROR
}
model ScenarioVariantCell {
id String @id @default(uuid()) @db.Uuid
inputHash String
output Json
statusCode Int
errorMessage String?
timeToComplete Int @default(0)
inputHash String? // TODO: Remove once migration is complete
output Json? // TODO: Remove once migration is complete
statusCode Int?
errorMessage String?
timeToComplete Int? @default(0) // TODO: Remove once migration is complete
retryTime DateTime?
streamingChannel String?
retrievalStatus CellRetrievalStatus @default(COMPLETE)
promptTokens Int? // Added promptTokens field
completionTokens Int? // Added completionTokens field
promptTokens Int? // TODO: Remove once migration is complete
completionTokens Int? // TODO: Remove once migration is complete
modelOutput ModelOutput?
promptVariantId String @db.Uuid
promptVariant PromptVariant @relation(fields: [promptVariantId], references: [id], onDelete: Cascade)
@@ -98,6 +109,24 @@ model ModelOutput {
updatedAt DateTime @updatedAt
@@unique([promptVariantId, testScenarioId])
}
model ModelOutput {
id String @id @default(uuid()) @db.Uuid
inputHash String
output Json
timeToComplete Int @default(0)
promptTokens Int?
completionTokens Int?
createdAt DateTime @default(now())
updatedAt DateTime @updatedAt
scenarioVariantCellId String @db.Uuid
scenarioVariantCell ScenarioVariantCell @relation(fields: [scenarioVariantCellId], references: [id], onDelete: Cascade)
@@unique([scenarioVariantCellId])
@@index([inputHash])
}

View File

@@ -16,7 +16,7 @@ const experiment = await prisma.experiment.create({
},
});
await prisma.modelOutput.deleteMany({
await prisma.scenarioVariantCell.deleteMany({
where: {
promptVariant: {
experimentId,