Paddle Multimodal Integration and eXploration, supporting mainstream multi-modal tasks, including end-to-end large-scale multi-modal pretrain models and diffusion model toolbox. Equipped with high performance and flexibility.
When deploying SD, it was found that there was an issue with the environment. It seems that there is a compatibility issue between Paddle and Paddlenlp. The error log is as follows:
Traceback (most recent call last): File "/data/dmx/sd_dir/ppdiffusers/examples/text_to_image_laion400m/train_txt2img_laion400m_trainer.py", line 19, in <module> from ldm import ( File "/data/dmx/sd_dir/ppdiffusers/examples/text_to_image_laion400m/ldm/__init__.py", line 16, in <module> from .ldm_args import DataArguments, ModelArguments, NoTrainerTrainingArguments File "/data/dmx/sd_dir/ppdiffusers/examples/text_to_image_laion400m/ldm/ldm_args.py", line 20, in <module> from paddlenlp.utils.log import logger File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/__init__.py", line 35, in <module> from . import ( File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/data/__init__.py", line 18, in <module> from .data_collator import * File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/data/data_collator.py", line 26, in <module> from ..transformers import BertTokenizer File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/transformers/__init__.py", line 17, in <module> from .model_utils import PretrainedModel, register_base_model File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/transformers/model_utils.py", line 62, in <module> from ..generation import GenerationConfig, GenerationMixin File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/generation/__init__.py", line 15, in <module> from .logits_process import ( File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/generation/logits_process.py", line 22, in <module> from paddle.nn.layer.layers import in_declarative_mode ModuleNotFoundError: No module named 'paddle.nn.layer.layers'
When deploying SD, it was found that there was an issue with the environment. It seems that there is a compatibility issue between Paddle and Paddlenlp. The error log is as follows:
Traceback (most recent call last): File "/data/dmx/sd_dir/ppdiffusers/examples/text_to_image_laion400m/train_txt2img_laion400m_trainer.py", line 19, in <module> from ldm import ( File "/data/dmx/sd_dir/ppdiffusers/examples/text_to_image_laion400m/ldm/__init__.py", line 16, in <module> from .ldm_args import DataArguments, ModelArguments, NoTrainerTrainingArguments File "/data/dmx/sd_dir/ppdiffusers/examples/text_to_image_laion400m/ldm/ldm_args.py", line 20, in <module> from paddlenlp.utils.log import logger File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/__init__.py", line 35, in <module> from . import ( File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/data/__init__.py", line 18, in <module> from .data_collator import * File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/data/data_collator.py", line 26, in <module> from ..transformers import BertTokenizer File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/transformers/__init__.py", line 17, in <module> from .model_utils import PretrainedModel, register_base_model File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/transformers/model_utils.py", line 62, in <module> from ..generation import GenerationConfig, GenerationMixin File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/generation/__init__.py", line 15, in <module> from .logits_process import ( File "/home/dmx/miniconda3/envs/sd/lib/python3.9/site-packages/paddlenlp/generation/logits_process.py", line 22, in <module> from paddle.nn.layer.layers import in_declarative_mode ModuleNotFoundError: No module named 'paddle.nn.layer.layers'