From f536038f651c2e02a5e6aba1963457e958c6c313 Mon Sep 17 00:00:00 2001 From: Ettore Di Giacinto Date: Fri, 17 Jan 2025 21:30:25 +0100 Subject: [PATCH] Do not always try to load the tokenizer Signed-off-by: Ettore Di Giacinto --- backend/python/transformers/backend.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/backend/python/transformers/backend.py b/backend/python/transformers/backend.py index 5459a078..9b65c6db 100644 --- a/backend/python/transformers/backend.py +++ b/backend/python/transformers/backend.py @@ -95,6 +95,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer): device_map="cpu" quantization = None + autoTokenizer = True if self.CUDA: from transformers import BitsAndBytesConfig, AutoModelForCausalLM @@ -198,9 +199,11 @@ class BackendServicer(backend_pb2_grpc.BackendServicer): device=device_map) self.OV = True elif request.Type == "MusicgenForConditionalGeneration": + autoTokenizer = False self.processor = AutoProcessor.from_pretrained(model_name) self.model = MusicgenForConditionalGeneration.from_pretrained(model_name) elif request.Type == "OuteTTS": + autoTokenizer = False options = request.Options MODELNAME = "OuteAI/OuteTTS-0.3-1B" TOKENIZER = "OuteAI/OuteTTS-0.3-1B" @@ -239,6 +242,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer): else: self.speaker = self.interface.load_default_speaker(name=SPEAKER) elif request.Type == "SentenceTransformer": + autoTokenizer = False self.model = SentenceTransformer(model_name, trust_remote_code=request.TrustRemoteCode) self.SentenceTransformer = True else: @@ -256,7 +260,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer): else: self.max_tokens = 512 - if request.Type != "MusicgenForConditionalGeneration": + if autoTokenizer: self.tokenizer = AutoTokenizer.from_pretrained(model_name, use_safetensors=True) self.XPU = False