Update app.py

app.py

@@ -74,173 +74,6 @@ clevr_model.to(device)
 clevr_model.load_state_dict(th.load(download_model('clevr_pos'), device))
 device = th.device('cpu' if not th.cuda.is_available() else 'cuda')
 
-# init stable diffusion model
-pipe = ComposableStableDiffusionPipeline.from_pretrained(
-    "CompVis/stable-diffusion-v1-4",
-).to(device)
-
-pipe.safety_checker = None
-
-# create model for CLEVR Objects
-clevr_options = model_and_diffusion_defaults_for_clevr()
-
-flags = {
-    "image_size": 128,
-    "num_channels": 192,
-    "num_res_blocks": 2,
-    "learn_sigma": True,
-    "use_scale_shift_norm": False,
-    "raw_unet": True,
-    "noise_schedule": "squaredcos_cap_v2",
-    "rescale_learned_sigmas": False,
-    "rescale_timesteps": False,
-    "num_classes": '2',
-    "dataset": "clevr_pos",
-    "use_fp16": has_cuda,
-    "timestep_respacing": '100'
-}
-
-for key, val in flags.items():
-    clevr_options[key] = val
-
-clevr_model, clevr_diffusion = create_model_and_diffusion_for_clevr(**clevr_options)
-clevr_model.eval()
-if has_cuda:
-    clevr_model.convert_to_fp16()
-
-clevr_model.to(device)
-clevr_model.load_state_dict(th.load(download_model('clevr_pos'), device))
-print('total clevr_pos parameters', sum(x.numel() for x in clevr_model.parameters()))
-
-print('creating base model...')
-base_name = 'base40M-textvec'
-base_model = model_from_config(MODEL_CONFIGS[base_name], device)
-base_model.eval()
-base_diffusion = diffusion_from_config(DIFFUSION_CONFIGS[base_name])
-
-print('creating upsample model...')
-upsampler_model = model_from_config(MODEL_CONFIGS['upsample'], device)
-upsampler_model.eval()
-upsampler_diffusion = diffusion_from_config(DIFFUSION_CONFIGS['upsample'])
-
-print('downloading base checkpoint...')
-base_model.load_state_dict(load_checkpoint(base_name, device))
-
-print('downloading upsampler checkpoint...')
-upsampler_model.load_state_dict(load_checkpoint('upsample', device))
-
-print('creating SDF model...')
-name = 'sdf'
-model = model_from_config(MODEL_CONFIGS[name], device)
-model.eval()
-
-print('loading SDF model...')
-model.load_state_dict(load_checkpoint(name, device))
-
-
-def compose_pointe(prompt, weights):
-    weight_list = [float(x.strip()) for x in weights.split('|')]
-    sampler = PointCloudSampler(
-        device=device,
-        models=[base_model, upsampler_model],
-        diffusions=[base_diffusion, upsampler_diffusion],
-        num_points=[1024, 4096 - 1024],
-        aux_channels=['R', 'G', 'B'],
-        guidance_scale=[weight_list, 0.0],
-        model_kwargs_key_filter=('texts', ''),  # Do not condition the upsampler at all
-    )
-
-    def generate_pcd(prompt_list):
-        # Produce a sample from the model.
-        samples = None
-        for x in tqdm(sampler.sample_batch_progressive(batch_size=1, model_kwargs=dict(texts=prompt_list))):
-            samples = x
-        return samples
-
-    def generate_fig(samples):
-        pc = sampler.output_to_point_clouds(samples)[0]
-        return pc
-
-
-# has_cuda = th.cuda.is_available()
-device = th.device('cpu' if not th.cuda.is_available() else 'cuda')
-
-# init stable diffusion model
-pipe = ComposableStableDiffusionPipeline.from_pretrained(
-    "CompVis/stable-diffusion-v1-4",
-).to(device)
-
-pipe.safety_checker = None
-
-# create model for CLEVR Objects
-clevr_options = model_and_diffusion_defaults_for_clevr()
-
-flags = {
-    "image_size": 128,
-    "num_channels": 192,
-    "num_res_blocks": 2,
-    "learn_sigma": True,
-    "use_scale_shift_norm": False,
-    "raw_unet": True,
-    "noise_schedule": "squaredcos_cap_v2",
-    "rescale_learned_sigmas": False,
-    "rescale_timesteps": False,
-    "num_classes": '2',
-    "dataset": "clevr_pos",
-    "use_fp16": has_cuda,
-    "timestep_respacing": '100'
-}
-
-for key, val in flags.items():
-    clevr_options[key] = val
-
-clevr_model, clevr_diffusion = create_model_and_diffusion_for_clevr(**clevr_options)
-clevr_model.eval()
-if has_cuda:
-    clevr_model.convert_to_fp16()
-
-clevr_model.to(device)
-clevr_model.load_state_dict(th.load(download_model('clevr_pos'), device))
-device = th.device('cpu' if not th.cuda.is_available() else 'cuda')
-
-# init stable diffusion model
-pipe = ComposableStableDiffusionPipeline.from_pretrained(
-    "CompVis/stable-diffusion-v1-4",
-).to(device)
-
-pipe.safety_checker = None
-
-# create model for CLEVR Objects
-clevr_options = model_and_diffusion_defaults_for_clevr()
-
-flags = {
-    "image_size": 128,
-    "num_channels": 192,
-    "num_res_blocks": 2,
-    "learn_sigma": True,
-    "use_scale_shift_norm": False,
-    "raw_unet": True,
-    "noise_schedule": "squaredcos_cap_v2",
-    "rescale_learned_sigmas": False,
-    "rescale_timesteps": False,
-    "num_classes": '2',
-    "dataset": "clevr_pos",
-    "use_fp16": has_cuda,
-    "timestep_respacing": '100'
-}
-
-for key, val in flags.items():
-    clevr_options[key] = val
-
-clevr_model, clevr_diffusion = create_model_and_diffusion_for_clevr(**clevr_options)
-clevr_model.eval()
-if has_cuda:
-    clevr_model.convert_to_fp16()
-
-clevr_model.to(device)
-clevr_model.load_state_dict(th.load(download_model('clevr_pos'), device))
-print('total clevr_pos parameters', sum(x.numel() for x in clevr_model.parameters()))
-
 print('creating base model...')
 base_name = 'base40M-textvec'
 base_model = model_from_config(MODEL_CONFIGS[base_name], device)