Revert "[LLM] IPEX auto importer turn on by default for XPU (#9730)" (#9759)

This reverts commit 0284801fbd.

python/llm/example/GPU/Deepspeed-AutoTP/deepspeed_autotp.py

@@ -25,6 +25,7 @@ world_size = int(os.getenv("WORLD_SIZE", "1"))
 from bigdl.llm import optimize_model
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Advanced-Quantizations/AWQ/generate.py

@@ -17,6 +17,7 @@
 import torch
 import time
 import argparse
+import intel_extension_for_pytorch as ipex
 from bigdl.llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 

python/llm/example/GPU/HF-Transformers-AutoModels/Advanced-Quantizations/GGUF/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Advanced-Quantizations/GPTQ/generate.py

@@ -17,6 +17,7 @@
 import torch
 import time
 import argparse
+import intel_extension_for_pytorch as ipex
 from bigdl.llm.transformers import AutoModelForCausalLM
 from transformers import LlamaTokenizer, GPTQConfig
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/aquila/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/aquila2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/baichuan/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/baichuan2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/bluelm/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/chatglm2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 import numpy as np

python/llm/example/GPU/HF-Transformers-AutoModels/Model/chatglm2/streamchat.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 import numpy as np

python/llm/example/GPU/HF-Transformers-AutoModels/Model/chatglm3/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 import numpy as np

python/llm/example/GPU/HF-Transformers-AutoModels/Model/chatglm3/streamchat.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 import numpy as np

python/llm/example/GPU/HF-Transformers-AutoModels/Model/chinese-llama2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/codellama/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/distil-whisper/recognize.py

@@ -14,6 +14,7 @@
 # limitations under the License.
 #
 
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/dolly-v1/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 import numpy as np

python/llm/example/GPU/HF-Transformers-AutoModels/Model/dolly-v2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 import numpy as np

python/llm/example/GPU/HF-Transformers-AutoModels/Model/falcon/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/flan-t5/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/gpt-j/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/internlm/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/llama2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/mistral/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/mixtral/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/mpt/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/phi-1_5/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 import numpy as np

python/llm/example/GPU/HF-Transformers-AutoModels/Model/qwen-vl/chat.py

@@ -22,6 +22,7 @@ from transformers import AutoTokenizer
 from transformers.generation import GenerationConfig
 
 from bigdl.llm.transformers import AutoModelForCausalLM
+import intel_extension_for_pytorch as ipex
 
 torch.manual_seed(1234)
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/qwen/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/replit/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/starcoder/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/vicuna/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/voiceassistant/generate.py

@@ -17,6 +17,7 @@
 import os
 import torch
 import time
+import intel_extension_for_pytorch as ipex
 import argparse
 import numpy as np
 import inquirer

python/llm/example/GPU/HF-Transformers-AutoModels/Model/whisper/recognize.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/Model/yi/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/HF-Transformers-AutoModels/More-Data-Types/transformers_low_bit_pipeline.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import argparse
 from bigdl.llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer, TextGenerationPipeline

python/llm/example/GPU/PyTorch-Models/Model/aquila2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/baichuan/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/baichuan2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/bluelm/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/chatglm2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/chatglm2/streamchat.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 import numpy as np

python/llm/example/GPU/PyTorch-Models/Model/chatglm3/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/chatglm3/streamchat.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 import numpy as np

python/llm/example/GPU/PyTorch-Models/Model/codellama/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/distil-whisper/recognize.py

@@ -14,6 +14,7 @@
 # limitations under the License.
 #
 
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/dolly-v1/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/dolly-v2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/flan-t5/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/llama2/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/llama2/low_memory_generate.py

@@ -23,6 +23,7 @@ import time
 import warnings
 from typing import Dict, List, Optional, Union
 
+import intel_extension_for_pytorch as ipex
 import torch
 import torch.nn as nn
 from accelerate import init_empty_weights

python/llm/example/GPU/PyTorch-Models/Model/llava/generate.py

@@ -34,6 +34,7 @@
 
 import argparse
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 
 from transformers import AutoModelForCausalLM

python/llm/example/GPU/PyTorch-Models/Model/mistral/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/mixtral/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/phi-1_5/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/qwen-vl/chat.py

@@ -22,6 +22,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 from transformers.generation import GenerationConfig
 
 from bigdl.llm import optimize_model
+import intel_extension_for_pytorch as ipex
 
 torch.manual_seed(1234)
 

python/llm/example/GPU/PyTorch-Models/Model/replit/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/starcoder/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Model/yi/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/More-Data-Types/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 

python/llm/example/GPU/PyTorch-Models/Save-Load/generate.py

@@ -15,6 +15,7 @@
 #
 
 import torch
+import intel_extension_for_pytorch as ipex
 import time
 import argparse
 from bigdl.llm import optimize_model

python/llm/example/GPU/QLoRA-FineTuning/alpaca-qlora/alpaca_qlora_finetuning.py

@@ -46,6 +46,7 @@ from peft import (
 )
 from utils.prompter import Prompter
 
+import intel_extension_for_pytorch as ipex
 from bigdl.llm.transformers import AutoModelForCausalLM
 # import them from bigdl.llm.transformers.qlora to get a BigDL-LLM compatible Peft model
 from bigdl.llm.transformers.qlora import get_peft_model, prepare_model_for_kbit_training,\

python/llm/example/GPU/QLoRA-FineTuning/qlora_finetuning.py

@@ -19,6 +19,7 @@ import os
 
 import transformers
 from transformers import LlamaTokenizer
+import intel_extension_for_pytorch as ipex
 from peft import LoraConfig
 from bigdl.llm.transformers.qlora import get_peft_model, prepare_model_for_kbit_training
 from bigdl.llm.transformers import AutoModelForCausalLM

python/llm/src/bigdl/llm/__init__.py

@@ -24,7 +24,7 @@ from .optimize import optimize_model
 import os
 
 # Default is false, set to true to auto importing Intel Extension for PyTorch.
-BIGDL_IMPORT_IPEX = os.getenv("BIGDL_IMPORT_IPEX", 'True').lower() in ('true', '1', 't')
+BIGDL_IMPORT_IPEX = os.getenv("BIGDL_IMPORT_IPEX", 'False').lower() in ('true', '1', 't')
 if BIGDL_IMPORT_IPEX:
     # Import Intel Extension for PyTorch as ipex if XPU version is installed
     from .utils.ipex_importer import ipex_importer

python/llm/src/bigdl/llm/utils/ipex_importer.py

@@ -35,19 +35,10 @@ class IPEXImporter:
         """
         # Check if xpu version installed
         try:
-            # Check if bigdl-core-xe is installed
             distribution('bigdl-core-xe')
             return True
         except PackageNotFoundError:
-            # bigdl-core-xe not found
-            # Check if bigdl-core-xe-21 is installed
-            try:
-                distribution('bigdl-core-xe-21')
-                return True
-            except PackageNotFoundError:
-                # bigdl-core-xe not found
-                return False
-        return False
+            return False
 
     def import_ipex(self):
         """