MCPcopy Create free account

hub / github.com/RameenAbdal/CLIP2StyleGAN / functions

Functions282 in github.com/RameenAbdal/CLIP2StyleGAN

⨍ Functions 282◇ Types & classes 64
↓ 15 callersMethodencode_image
(self, image)
clipmod/model.py:336
↓ 15 callersFunctionupdate
(state_dict, new)
edits_sg/convert_weight.py:95
↓ 15 callersFunctionupfirdn2d
(input, kernel, up=1, down=1, pad=(0, 0))
edits_sg/op/upfirdn2d.py:149
↓ 14 callersMethod__init__
(self, kernel, pad, upsample_factor=1)
edits_sg/model.py:76
↓ 13 callersFunctionrandom_mat_apply
(p, transform, prev, eye, device="cpu")
edits_sg/non_leaking.py:185
↓ 8 callersMethodencode_text
(self, text)
clipmod/model.py:339
↓ 6 callersMethod__init__
(self, in_channel, style_dim, upsample=True, blur_kernel=[1, 3, 3, 1])
edits_sg/swagan.py:120
↓ 6 callersMethod__init__
(self, embed_dim: int, # vision image_resolution: int,
clipmod/model.py:240
↓ 6 callersMethodencode
(self, text)
clipmod/simple_tokenizer.py:121
↓ 5 callersMethod__init__
(self, pnet_type='vgg', pnet_rand=False, pnet_tune=False, use_dropout=True, spatial=False, version='0.1', lpip
edits_sg/lpips/networks_basic.py:28
↓ 5 callersFunctionconvert_conv
(vars, source_name, target_name, bias=True, start=0)
edits_sg/convert_weight.py:42
↓ 5 callersMethodsave
(self, path, label)
edits_sg/lpips/dist_model.py:195
↓ 4 callersMethod__init__
Build pretrained InceptionV3 Parameters ---------- output_blocks : list of int Indices of blocks to return featur
edits_sg/inception.py:31
↓ 4 callersMethod_make_layer
(self, planes, blocks, stride=1)
clipmod/model.py:126
↓ 4 callersMethodbackward
(ctx, grad_output)
edits_sg/op/upfirdn2d.py:128
↓ 4 callersFunctionconv2d_gradfix
( transpose, weight_shape, stride, padding, output_padding, dilation, groups )
edits_sg/op/conv2d_gradfix.py:104
↓ 4 callersFunctionensure_tuple
(xs, ndim)
edits_sg/op/conv2d_gradfix.py:95
↓ 4 callersMethodforward
(self, in0, in1, retPerLayer=None)
edits_sg/lpips/networks_basic.py:152
↓ 4 callersFunctionlognormal_sample
(size, mean=0, std=1, device="cpu")
edits_sg/non_leaking.py:162
↓ 4 callersMethodmean_latent
(self, n_latent)
edits_sg/model.py:488
↓ 4 callersFunctionnormalize
(x)
edits_sg/ppl.py:12
↓ 4 callersFunctionscale_mat_single
(s_x, s_y)
edits_sg/non_leaking.py:380
↓ 4 callersFunctionuniform_sample
(size, low, high, device="cpu")
edits_sg/non_leaking.py:173
↓ 3 callersMethod__init__
(self, requires_grad=False, pretrained=True)
edits_sg/lpips/pretrained_networks.py:98
↓ 3 callersFunctioncategory_sample
(size, categories, device="cpu")
edits_sg/non_leaking.py:166
↓ 3 callersFunctionconvert_dense
(vars, source_name, target_name)
edits_sg/convert_weight.py:81
↓ 3 callersFunctionconvert_modconv
(vars, source_name, target_name, flip=False)
edits_sg/convert_weight.py:14
↓ 3 callersMethoddecode
(self, tokens)
clipmod/simple_tokenizer.py:129
↓ 3 callersMethodforward
Function computes the distance between image patches in0 and in1 INPUTS in0, in1 - torch.Tensor object of shape Nx3xXxY - image p
edits_sg/lpips/dist_model.py:109
↓ 3 callersFunctionget_world_size
()
edits_sg/distributed.py:34
↓ 3 callersFunctionmake_kernel
(k)
edits_sg/model.py:22
↓ 3 callersFunctionrotate_mat
(theta, device="cpu")
edits_sg/non_leaking.py:77
↓ 3 callersMethodsave_network
(self, network, path, network_label, epoch_label)
edits_sg/lpips/base_model.py:38
↓ 3 callersFunctionscale_mat
(s_x, s_y, device="cpu")
edits_sg/non_leaking.py:89
↓ 3 callersFunctiontranslate_mat_single
(t_x, t_y)
edits_sg/non_leaking.py:384
↓ 2 callersFunction_transform
(n_px)
clipmod/clip.py:72
↓ 2 callersFunctionbytes_to_unicode
Returns list of utf-8 byte and a corresponding list of unicode strings. The reversible bpe codes work on unicode strings. This means you
clipmod/simple_tokenizer.py:16
↓ 2 callersFunctioncalc_output_padding
(input_shape, output_shape)
edits_sg/op/conv2d_gradfix.py:122
↓ 2 callersFunctionconvert_torgb
(vars, source_name, target_name)
edits_sg/convert_weight.py:60
↓ 2 callersFunctioncould_use_op
(input)
edits_sg/op/conv2d_gradfix.py:78
↓ 2 callersFunctionfill_statedict
(state_dict, vars, size, n_mlp)
edits_sg/convert_weight.py:140
↓ 2 callersFunctionfused_leaky_relu
(input, bias=None, negative_slope=0.2, scale=2 ** 0.5)
edits_sg/op/fused_act.py:110
↓ 2 callersFunctionget_haar_wavelet
(in_channels)
edits_sg/swagan.py:25
↓ 2 callersMethodget_latent
(self, input)
edits_sg/model.py:496
↓ 2 callersFunctionget_pairs
Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).
clipmod/simple_tokenizer.py:38
↓ 2 callersFunctionim2tensor
(image, imtype=np.uint8, cent=1., factor=255./2.)
edits_sg/lpips/__init__.py:111
↓ 2 callersFunctionlerp
(a, b, t)
edits_sg/ppl.py:27
↓ 2 callersFunctionload
Load a CLIP model Parameters ---------- name : str A model name listed by `clip.available_models()`, or the path to a model check
clipmod/clip.py:87
↓ 2 callersFunctionload_patched_inception_v3
()
edits_sg/calc_inception.py:51
↓ 2 callersMethodmake_noise
(self)
edits_sg/model.py:477
↓ 2 callersFunctionnormal_sample
(size, mean=0, std=1, device="cpu")
edits_sg/non_leaking.py:177
↓ 2 callersFunctionnp2tensor
(np_obj)
edits_sg/lpips/__init__.py:66
↓ 2 callersFunctionpatch_device
(module)
clipmod/clip.py:140
↓ 2 callersFunctionpatch_float
(module)
clipmod/clip.py:164
↓ 2 callersMethodsave
(self, label)
edits_sg/lpips/base_model.py:34
↓ 2 callersFunctionspatial_average
(in_tens, keepdim=True)
edits_sg/lpips/networks_basic.py:17
↓ 2 callersFunctiontranslate_mat
(t_x, t_y, device="cpu")
edits_sg/non_leaking.py:67
↓ 2 callersFunctionupsample
(in_tens, out_H=64)
edits_sg/lpips/networks_basic.py:20
↓ 1 callersFunction_download
(url: str, root: str)
clipmod/clip.py:40
↓ 1 callersFunctionaccumulate
(model1, model2, decay=0.999)
edits_sg/edit_cars.py:13
↓ 1 callersFunctionaccumulate
(model1, model2, decay=0.999)
edits_sg/edit_faces.py:12
↓ 1 callersFunctionapply_color
(img, mat)
edits_sg/non_leaking.py:441
↓ 1 callersMethodattention
(self, x: torch.Tensor)
clipmod/model.py:181
↓ 1 callersFunctionavailable_models
Returns the names of available CLIP models
clipmod/clip.py:82
↓ 1 callersMethodbackward_train
(self)
edits_sg/lpips/dist_model.py:162
↓ 1 callersFunctionbasic_clean
(text)
clipmod/simple_tokenizer.py:50
↓ 1 callersFunctionbernoulli_sample
(size, p, device="cpu")
edits_sg/non_leaking.py:181
↓ 1 callersMethodbpe
(self, token)
clipmod/simple_tokenizer.py:80
↓ 1 callersMethodbuild_attention_mask
(self)
clipmod/model.py:324
↓ 1 callersFunctionbuild_model
(state_dict: dict)
clipmod/model.py:459
↓ 1 callersFunctioncalc_fid
(sample_mean, sample_cov, real_mean, real_cov, eps=1e-6)
edits_sg/fid.py:34
↓ 1 callersMethodclamp_weights
(self)
edits_sg/lpips/dist_model.py:127
↓ 1 callersMethodcompute_accuracy
d0, d1 are Variables, judge is a Tensor
edits_sg/lpips/dist_model.py:165
↓ 1 callersFunctioncompute_scores
()
test_disentanglement.py:10
↓ 1 callersFunctioncompute_scores
()
test_clip_scores.py:14
↓ 1 callersFunctionconvert_weights
Convert applicable model parameters to fp16
clipmod/model.py:435
↓ 1 callersFunctiondefault_bpe
()
clipmod/simple_tokenizer.py:11
↓ 1 callersFunctiondiscriminator_fill_statedict
(statedict, vars, size)
edits_sg/convert_weight.py:106
↓ 1 callersFunctionedit
(gma)
edits_sg/edit_cars.py:21
↓ 1 callersFunctionedit
(g_ema)
edits_sg/edit_faces.py:19
↓ 1 callersFunctionextract_feature_from_samples
( generator, inception, truncation, truncation_latent, batch_size, n_sample, device )
edits_sg/fid.py:15
↓ 1 callersFunctionextract_features
(loader, inception, device)
edits_sg/calc_inception.py:61
↓ 1 callersFunctionfid_inception_v3
Build pretrained Inception model for FID computation The Inception model for FID computation uses a different set of weights and has a slight
edits_sg/inception.py:166
↓ 1 callersMethodforward
(self, image, text)
clipmod/model.py:418
↓ 1 callersMethodforward_train
(self)
edits_sg/lpips/dist_model.py:148
↓ 1 callersFunctiongenerate_pca_images
( image_features)
compute_pca.py:17
↓ 1 callersFunctionget_lr
(t, initial_lr, rampdown=0.25, rampup=0.05)
edits_sg/projector.py:47
↓ 1 callersFunctionget_padding
(G, height, width, kernel_size)
edits_sg/non_leaking.py:301
↓ 1 callersMethodinitialize
INPUTS model - ['net-lin'] for linearly calibrated network ['net'] for off-the-shelf network
edits_sg/lpips/dist_model.py:28
↓ 1 callersMethodinitialize_parameters
(self)
clipmod/model.py:295
↓ 1 callersFunctionlatent_noise
(latent, strength)
edits_sg/projector.py:55
↓ 1 callersFunctionluma_flip_mat
(axis, i)
edits_sg/non_leaking.py:141
↓ 1 callersFunctionmake_image
(tensor)
edits_sg/projector.py:61
↓ 1 callersMethodname
(self)
edits_sg/lpips/dist_model.py:25
↓ 1 callersFunctionnoise_normalize_
(noises)
edits_sg/projector.py:39
↓ 1 callersFunctionnoise_regularize
(noises)
edits_sg/projector.py:16
↓ 1 callersMethodoptimize_text
(self, optim_var)
clipmod/model.py:365
↓ 1 callersFunctionprepare
( env, dataset, n_worker, sizes=(128, 256, 512, 1024), resample=Image.LANCZOS )
edits_sg/prepare_data.py:43
↓ 1 callersFunctionrandom_apply_affine
(img, p, G=None, antialiasing_kernel=SYM6)
edits_sg/non_leaking.py:388
↓ 1 callersFunctionrandom_apply_color
(img, p, C=None)
edits_sg/non_leaking.py:452
next →1–100 of 282, ranked by callers