This commit is contained in:
Timothy J. Baek 2024-04-10 00:33:45 -07:00
parent 48aad65514
commit f4b87ecb23
3 changed files with 243 additions and 236 deletions

View file

@ -44,7 +44,7 @@ from apps.web.models.documents import (
DocumentResponse, DocumentResponse,
) )
from apps.rag.utils import query_doc, query_collection, embedding_model_get_path from apps.rag.utils import query_doc, query_collection, get_embedding_model_path
from utils.misc import ( from utils.misc import (
calculate_sha256, calculate_sha256,
@ -77,10 +77,14 @@ app.state.PDF_EXTRACT_IMAGES = False
app.state.CHUNK_SIZE = CHUNK_SIZE app.state.CHUNK_SIZE = CHUNK_SIZE
app.state.CHUNK_OVERLAP = CHUNK_OVERLAP app.state.CHUNK_OVERLAP = CHUNK_OVERLAP
app.state.RAG_TEMPLATE = RAG_TEMPLATE app.state.RAG_TEMPLATE = RAG_TEMPLATE
app.state.RAG_EMBEDDING_MODEL = RAG_EMBEDDING_MODEL app.state.RAG_EMBEDDING_MODEL = RAG_EMBEDDING_MODEL
app.state.RAG_EMBEDDING_MODEL_PATH = embedding_model_get_path( app.state.RAG_EMBEDDING_MODEL_PATH = get_embedding_model_path(
app.state.RAG_EMBEDDING_MODEL, RAG_EMBEDDING_MODEL_AUTO_UPDATE app.state.RAG_EMBEDDING_MODEL, RAG_EMBEDDING_MODEL_AUTO_UPDATE
) )
app.state.TOP_K = 4 app.state.TOP_K = 4
app.state.sentence_transformer_ef = ( app.state.sentence_transformer_ef = (
@ -148,7 +152,7 @@ async def update_embedding_model(
) )
try: try:
app.state.RAG_EMBEDDING_MODEL_PATH = embedding_model_get_path( app.state.RAG_EMBEDDING_MODEL_PATH = get_embedding_model_path(
app.state.RAG_EMBEDDING_MODEL, True app.state.RAG_EMBEDDING_MODEL, True
) )
app.state.sentence_transformer_ef = ( app.state.sentence_transformer_ef = (

View file

@ -192,21 +192,21 @@ def rag_messages(docs, messages, template, k, embedding_function):
return messages return messages
def embedding_model_get_path( def get_embedding_model_path(
embedding_model: str, update_embedding_model: bool = False embedding_model: str, update_embedding_model: bool = False
): ):
# Construct huggingface_hub kwargs with local_files_only to return the snapshot path # Construct huggingface_hub kwargs with local_files_only to return the snapshot path
cache_dir = os.getenv("SENTENCE_TRANSFORMERS_HOME") cache_dir = os.getenv("SENTENCE_TRANSFORMERS_HOME")
local_files_only = not update_embedding_model local_files_only = not update_embedding_model
snapshot_kwargs = { snapshot_kwargs = {
"cache_dir": cache_dir, "cache_dir": cache_dir,
"local_files_only": local_files_only, "local_files_only": local_files_only,
} }
log.debug(f"SENTENCE_TRANSFORMERS_HOME cache_dir: {cache_dir}")
log.debug(f"embedding_model: {embedding_model}") log.debug(f"embedding_model: {embedding_model}")
log.debug(f"update_embedding_model: {update_embedding_model}") log.debug(f"snapshot_kwargs: {snapshot_kwargs}")
log.debug(f"local_files_only: {local_files_only}")
# Inspiration from upstream sentence_transformers # Inspiration from upstream sentence_transformers
if ( if (

View file

@ -21,8 +21,8 @@
export let saveHandler: Function; export let saveHandler: Function;
let loading = false; let scanDirLoading = false;
let loading1 = false; let updateEmbeddingModelLoading = false;
let showResetConfirm = false; let showResetConfirm = false;
@ -35,14 +35,12 @@
k: 4 k: 4
}; };
let embeddingModel = { let embeddingModel = '';
embedding_model: ''
};
const scanHandler = async () => { const scanHandler = async () => {
loading = true; scanDirLoading = true;
const res = await scanDocs(localStorage.token); const res = await scanDocs(localStorage.token);
loading = false; scanDirLoading = false;
if (res) { if (res) {
await documents.set(await getDocs(localStorage.token)); await documents.set(await getDocs(localStorage.token));
@ -51,7 +49,7 @@
}; };
const embeddingModelUpdateHandler = async () => { const embeddingModelUpdateHandler = async () => {
if (embeddingModel.embedding_model.split('/').length - 1 > 1) { if (embeddingModel.split('/').length - 1 > 1) {
toast.error( toast.error(
$i18n.t( $i18n.t(
'Model filesystem path detected. Model shortname is required for update, cannot continue.' 'Model filesystem path detected. Model shortname is required for update, cannot continue.'
@ -60,11 +58,11 @@
return; return;
} }
console.log('Update embedding model attempt:', embeddingModel.embedding_model); console.log('Update embedding model attempt:', embeddingModel);
loading1 = true; updateEmbeddingModelLoading = true;
const res = await updateEmbeddingModel(localStorage.token, embeddingModel); const res = await updateEmbeddingModel(localStorage.token, { embedding_model: embeddingModel });
loading1 = false; updateEmbeddingModelLoading = false;
if (res) { if (res) {
console.log('embeddingModelUpdateHandler:', res); console.log('embeddingModelUpdateHandler:', res);
@ -101,7 +99,9 @@
chunkOverlap = res.chunk.chunk_overlap; chunkOverlap = res.chunk.chunk_overlap;
} }
embeddingModel = await getEmbeddingModel(localStorage.token); const embeddingModelConfig = await getEmbeddingModel(localStorage.token);
embeddingModel = embeddingModelConfig.embedding_model;
querySettings = await getQuerySettings(localStorage.token); querySettings = await getQuerySettings(localStorage.token);
}); });
@ -124,7 +124,7 @@
</div> </div>
<button <button
class=" self-center text-xs p-1 px-3 bg-gray-100 dark:bg-gray-800 dark:hover:bg-gray-700 rounded flex flex-row space-x-1 items-center {loading class=" self-center text-xs p-1 px-3 bg-gray-100 dark:bg-gray-800 dark:hover:bg-gray-700 rounded flex flex-row space-x-1 items-center {scanDirLoading
? ' cursor-not-allowed' ? ' cursor-not-allowed'
: ''}" : ''}"
on:click={() => { on:click={() => {
@ -132,115 +132,114 @@
console.log('check'); console.log('check');
}} }}
type="button" type="button"
disabled={loading} disabled={scanDirLoading}
> >
<div class="self-center font-medium">{$i18n.t('Scan')}</div> <div class="self-center font-medium">{$i18n.t('Scan')}</div>
<!-- <svg {#if scanDirLoading}
<div class="ml-3 self-center">
<svg
class=" w-3 h-3"
viewBox="0 0 24 24"
fill="currentColor"
xmlns="http://www.w3.org/2000/svg"
><style>
.spinner_ajPY {
transform-origin: center;
animation: spinner_AtaB 0.75s infinite linear;
}
@keyframes spinner_AtaB {
100% {
transform: rotate(360deg);
}
}
</style><path
d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
opacity=".25"
/><path
d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
class="spinner_ajPY"
/></svg
>
</div>
{/if}
</button>
</div>
</div>
<hr class=" dark:border-gray-700" />
<div class="space-y-2">
<div>
<div class=" mb-2 text-sm font-medium">{$i18n.t('Update Embedding Model')}</div>
<div class="flex w-full">
<div class="flex-1 mr-2">
<input
class="w-full rounded-lg py-2 px-4 text-sm dark:text-gray-300 dark:bg-gray-850 outline-none"
placeholder={$i18n.t('Update embedding model (e.g. {{model}})', {
model: embeddingModel.slice(-40)
})}
bind:value={embeddingModel}
/>
</div>
<button
class="px-2.5 bg-gray-100 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg transition"
on:click={() => {
embeddingModelUpdateHandler();
}}
disabled={updateEmbeddingModelLoading}
>
{#if updateEmbeddingModelLoading}
<div class="self-center">
<svg
class=" w-4 h-4"
viewBox="0 0 24 24"
fill="currentColor"
xmlns="http://www.w3.org/2000/svg"
><style>
.spinner_ajPY {
transform-origin: center;
animation: spinner_AtaB 0.75s infinite linear;
}
@keyframes spinner_AtaB {
100% {
transform: rotate(360deg);
}
}
</style><path
d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
opacity=".25"
/><path
d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
class="spinner_ajPY"
/></svg
>
</div>
{:else}
<svg
xmlns="http://www.w3.org/2000/svg" xmlns="http://www.w3.org/2000/svg"
viewBox="0 0 16 16" viewBox="0 0 16 16"
fill="currentColor" fill="currentColor"
class="w-3 h-3" class="w-4 h-4"
> >
<path <path
fill-rule="evenodd" d="M8.75 2.75a.75.75 0 0 0-1.5 0v5.69L5.03 6.22a.75.75 0 0 0-1.06 1.06l3.5 3.5a.75.75 0 0 0 1.06 0l3.5-3.5a.75.75 0 0 0-1.06-1.06L8.75 8.44V2.75Z"
d="M13.836 2.477a.75.75 0 0 1 .75.75v3.182a.75.75 0 0 1-.75.75h-3.182a.75.75 0 0 1 0-1.5h1.37l-.84-.841a4.5 4.5 0 0 0-7.08.932.75.75 0 0 1-1.3-.75 6 6 0 0 1 9.44-1.242l.842.84V3.227a.75.75 0 0 1 .75-.75Zm-.911 7.5A.75.75 0 0 1 13.199 11a6 6 0 0 1-9.44 1.241l-.84-.84v1.371a.75.75 0 0 1-1.5 0V9.591a.75.75 0 0 1 .75-.75H5.35a.75.75 0 0 1 0 1.5H3.98l.841.841a4.5 4.5 0 0 0 7.08-.932.75.75 0 0 1 1.025-.273Z"
clip-rule="evenodd"
/> />
</svg> --> <path
d="M3.5 9.75a.75.75 0 0 0-1.5 0v1.5A2.75 2.75 0 0 0 4.75 14h6.5A2.75 2.75 0 0 0 14 11.25v-1.5a.75.75 0 0 0-1.5 0v1.5c0 .69-.56 1.25-1.25 1.25h-6.5c-.69 0-1.25-.56-1.25-1.25v-1.5Z"
{#if loading} />
<div class="ml-3 self-center"> </svg>
<svg
class=" w-3 h-3"
viewBox="0 0 24 24"
fill="currentColor"
xmlns="http://www.w3.org/2000/svg"
><style>
.spinner_ajPY {
transform-origin: center;
animation: spinner_AtaB 0.75s infinite linear;
}
@keyframes spinner_AtaB {
100% {
transform: rotate(360deg);
}
}
</style><path
d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
opacity=".25"
/><path
d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
class="spinner_ajPY"
/></svg
>
</div>
{/if} {/if}
</button> </button>
</div> </div>
</div>
<hr class=" dark:border-gray-700" /> <div class="mt-2 mb-1 text-xs text-gray-400 dark:text-gray-500">
{$i18n.t(
<div> 'Warning: If you update or change your embedding model, you will need to re-import all documents.'
<div class=" flex w-full justify-between">
<Tooltip content={$i18n.t('Embedding model: {{embedding_model}}', embeddingModel)}>
<div class=" self-center text-xs font-medium">
{$i18n.t('Update embedding model {{embedding_model}}', {
embedding_model: embeddingModel.embedding_model.slice(-40)
})}
</div>
</Tooltip>
<Tooltip
content={$i18n.t(
'Understand that updating or changing your embedding model requires reset of the vector database and re-import of all documents. You have been warned!'
)} )}
>
<button
class=" self-center text-xs p-1 px-3 bg-gray-100 dark:bg-gray-800 dark:hover:bg-gray-700 rounded flex flex-row space-x-1 items-center {loading1
? ' cursor-not-allowed'
: ''}"
on:click={() => {
embeddingModelUpdateHandler(embeddingModel);
}}
type="button"
disabled={loading1}
>
<div class="self-center font-medium">{$i18n.t('Update')}</div>
{#if loading1}
<div class="ml-3 self-center">
<svg
class=" w-3 h-3"
viewBox="0 0 24 24"
fill="currentColor"
xmlns="http://www.w3.org/2000/svg"
><style>
.spinner_ajPY {
transform-origin: center;
animation: spinner_AtaB 0.75s infinite linear;
}
@keyframes spinner_AtaB {
100% {
transform: rotate(360deg);
}
}
</style><path
d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
opacity=".25"
/><path
d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
class="spinner_ajPY"
/></svg
>
</div>
{/if}
</button>
</Tooltip>
</div>
</div> </div>
<hr class=" dark:border-gray-700" /> <hr class=" dark:border-gray-700 my-3" />
<div class=" "> <div class=" ">
<div class=" text-sm font-medium">{$i18n.t('Chunk Params')}</div> <div class=" text-sm font-medium">{$i18n.t('Chunk Params')}</div>
@ -262,7 +261,9 @@
</div> </div>
<div class="flex w-full"> <div class="flex w-full">
<div class=" self-center text-xs font-medium min-w-fit">{$i18n.t('Chunk Overlap')}</div> <div class=" self-center text-xs font-medium min-w-fit">
{$i18n.t('Chunk Overlap')}
</div>
<div class="self-center p-3"> <div class="self-center p-3">
<input <input
@ -440,4 +441,6 @@
{$i18n.t('Save')} {$i18n.t('Save')}
</button> </button>
</div> </div>
</div>
</div>
</form> </form>