inference using custom diffusion

[rezkanas]

I train a model on my photos using custom diffusion with below scripts...

!accelerate launch train_custom_diffusion.py \
--pretrained_model_name_or_path=$MODEL_NAME  \
--instance_data_dir=$INSTANCE_DIR \
--output_dir=$OUTPUT_DIR \
--class_data_dir=$class_data_dir \
--with_prior_preservation --prior_loss_weight=1.0 \
--class_prompt="photo of a person" --num_class_images=200 \
--instance_prompt="photo of a <new1> person"  \
--resolution=512  \
--train_batch_size=2  \
--learning_rate=5e-6  \
--lr_warmup_steps=0 \
--max_train_steps=1200 \
--scale_lr --hflip --noaug \
--modifier_token "<new1>"\
--enable_xformers_memory_efficient_attention \
--no_safe_serialization

first inconvenience, that i could not use '--freeze_model=crossattn ' recommended by README file when training faces, moreover, I could not save my final model as per the issue

but I have saved some checkpoints that I am trying to use for inference with this code...

pipe = DiffusionPipeline.from_pretrained(
    "CompVis/stable-diffusion-v1-4", torch_dtype=torch.float16
).to("cuda")
pipe.unet.load_attn_procs(
    "/mnt/c/Users/noobw/PycharmProjects/pythonProject/custom_diffusion_anas/checkpoint-1000"
)
pipe.load_textual_inversion("/mnt/c/Users/noobw/PycharmProjects/pythonProject/custom_diffusion_anas/checkpoint-1000", weight_name="<new1>.bin")

image = pipe(
    "<new1> person sitting in a bucket",
    num_inference_steps=100,
    guidance_scale=6.0,
    eta=1.0,
).images[0]

but I am receiving an error

'OSError: We couldn't connect to 'https://huggingface.co' to load this model, couldn't find it in the cached files and it looks like /mnt/c/Users/noobw/PycharmProjects/pythonProject/custom_diffusion_anas/checkpoint-1000 is not the path to a directory containing a file named pytorch_lora_weights.bin or
Checkout your internet connection or see how to run the library in offline mode at 'https://huggingface.co/docs/diffusers/installation#offline-mode'.'

I went to the direcotry and the files there are the following:

how do you think I should change my code? honestly I could not tell which weights i need to load for unet and textual inversion... can someone here help ?

@levi @kig @kashif