Open Revanthraja opened 1 year ago
Hi @Revanthraja. Please use the inference.py script.
python inference.py --model '/path/to/checkpoint/' --prompt 'apple' -n "orange"
The usage are written in the readme file.
Hello sir in which folder I will get this path '/path/to/checkpoint/' please help me sir
You should have executed a command like the following:
python train.py --config configs/my_train_config.yaml
The "output_dir" in config.yaml used by your training, is where checkpoints are saved. Please check that directory.
like I need use like this !python inference.py \ --model /content/Text-To-Video-Finetuning/models/model_scope_diffusers\ --prompt "The two persons are playing the cricket" \ --num-frames 30 \ --window-size 12 \ --width 1024 \ --height 576 \ --sdp
Hello sir then how to pass the video and text directory please help in this
%cd /content/Text-To-Video-Finetuning
model = "/content/outputs/train_2023-07-13T11-37-44" #@param {type:"string"} prompt = "Cricket playing" #@param {type:"string"} negative = "text, watermark, copyright, blurry, low resolution, blur, low quality" #@param {type:"string"} prompt = f"\"{prompt}\"" negative = f"\"{negative}\"" num_steps = 25 #@param {type:"raw"} guidance_scale = 23 #@param {type:"raw"} fps = 24 #@param {type:"raw"} num_frames = 10 #@param {type:"raw"} !python inference.py -m {model} -p {prompt} -n {negative} -W 1024 -H 576 -o /content/video_outputs -d cuda -x -s {num_steps} -g {guidance_scale} -f {fps} -T {num_frames} when I excecuting this code I am getting OSError: Error no file named scheduler_config.json found in directory /content/outputs/train_2023-07-13T11-37-44. this error please any one help me in this
Is there a 'scheduler' directory in the 'train_2023-07-13T11-37-44' ? That directory must contain 'scheduler_config.json', but it probably disappeared for any reason.
Hello sir can you tell me for which reason it is disappeared for trainig I used low_vram_config file
Steps: 0% 0/100 [00:00<?, ?it/s]2200 params have been unfrozen for training.
/usr/local/lib/python3.10/dist-packages/diffusers/configuration_utils.py:134: FutureWarning: Accessing config attribute num_train_timesteps
directly via 'DDPMScheduler' object attribute is deprecated. Please access 'num_train_timesteps' over 'DDPMScheduler's config object instead, e.g. 'scheduler.config.num_train_timesteps'.
deprecate("direct config name access", "1.0.0", deprecation_message, standard_warn=False)
/usr/local/lib/python3.10/dist-packages/diffusers/configuration_utils.py:134: FutureWarning: Accessing config attribute prediction_type
directly via 'DDPMScheduler' object attribute is deprecated. Please access 'prediction_type' over 'DDPMScheduler's config object instead, e.g. 'scheduler.config.prediction_type'.
deprecate("direct config name access", "1.0.0", deprecation_message, standard_warn=False)
Steps: 100% 100/100 [01:12<00:00, 1.84it/s, lr=1e-6, step_loss=0.00567]Traceback (most recent call last):
File "/content/Text-To-Video-Finetuning/train.py", line 999, in
I am getting this error in google colab plz anyone help me
Hello sir after training the model then how to test my model giving text as input please help me in this issue