first experiment

args.json

@@ -0,0 +1,210 @@
+{
+  "pretrained_model_name_or_path": "runwayml/stable-diffusion-v1-5",
+  "pretrained_vae_name_or_path": "stabilityai/sd-vae-ft-mse",
+  "revision": "fp16",
+  "tokenizer_name": null,
+  "instance_data_dir": null,
+  "class_data_dir": null,
+  "instance_prompt": null,
+  "class_prompt": null,
+  "save_sample_prompt": null,
+  "save_sample_negative_prompt": null,
+  "n_save_sample": 4,
+  "save_guidance_scale": 7.5,
+  "save_infer_steps": 20,
+  "pad_tokens": false,
+  "with_prior_preservation": true,
+  "prior_loss_weight": 1.0,
+  "num_class_images": 50,
+  "output_dir": "/content/stable_diffusion_weights/model_herman",
+  "seed": 1337,
+  "resolution": 512,
+  "center_crop": false,
+  "train_text_encoder": true,
+  "train_batch_size": 1,
+  "sample_batch_size": 4,
+  "num_train_epochs": 1,
+  "max_train_steps": 800,
+  "gradient_accumulation_steps": 1,
+  "gradient_checkpointing": false,
+  "learning_rate": 1e-06,
+  "scale_lr": false,
+  "lr_scheduler": "constant",
+  "lr_warmup_steps": 0,
+  "use_8bit_adam": true,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.999,
+  "adam_weight_decay": 0.01,
+  "adam_epsilon": 1e-08,
+  "max_grad_norm": 1.0,
+  "push_to_hub": true,
+  "hub_token": "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",
+  "hub_model_id": null,
+  "logging_dir": "logs",
+  "log_interval": 10,
+  "save_interval": 10000,
+  "save_min_steps": 0,
+  "mixed_precision": "fp16",
+  "not_cache_latents": false,
+  "hflip": false,
+  "local_rank": -1,
+  "concepts_list": [
+    {
+      "instance_prompt": "photo of Baiturrahman Grand Mosque in Aceh",
+      "class_prompt": "photo of a Baiturrahman Grand Mosque in Aceh",
+      "instance_data_dir": "/content/data/Baiturrahman Grand Mosque - Aceh",
+      "class_data_dir": "/content/data_class/Baiturrahman Grand Mosque - Aceh"
+    },
+    {
+      "instance_prompt": "photo of Benteng Van Den Bosch in Ngawi",
+      "class_prompt": "photo of a Benteng Van Den Bosch in Ngawi",
+      "instance_data_dir": "/content/data/Benteng Van Den Bosch - Ngawi",
+      "class_data_dir": "/content/data_class/Benteng Van Den Bosch - Ngawi"
+    },
+    {
+      "instance_prompt": "photo of Benteng Vredeburg in Yogyakarta",
+      "class_prompt": "photo of a Benteng Vredeburg in Yogyakarta",
+      "instance_data_dir": "/content/data/Benteng Vredeburg - Yogyakarta",
+      "class_data_dir": "/content/data_class/Benteng Vredeburg - Yogyakarta"
+    },
+    {
+      "instance_prompt": "photo of Bukit Tinggi in Sumatera Barat",
+      "class_prompt": "photo of a Bukit Tinggi in Sumatera Barat",
+      "instance_data_dir": "/content/data/Bukit Tinggi - Sumatera Barat",
+      "class_data_dir": "/content/data_class/Bukit Tinggi - Sumatera Barat"
+    },
+    {
+      "instance_prompt": "photo of Candi Borobudur in Magelang",
+      "class_prompt": "photo of a Candi Borobudur in Magelang",
+      "instance_data_dir": "/content/data/Candi Borobudur - Magelang",
+      "class_data_dir": "/content/data_class/Candi Borobudur - Magelang"
+    },
+    {
+      "instance_prompt": "photo of Candi Brahu in Mojokerto",
+      "class_prompt": "photo of a Candi Brahu in Mojokerto",
+      "instance_data_dir": "/content/data/Candi Brahu - Mojokerto",
+      "class_data_dir": "/content/data_class/Candi Brahu - Mojokerto"
+    },
+    {
+      "instance_prompt": "photo of Candi Kalasan in Yogyakarta",
+      "class_prompt": "photo of a Candi Kalasan in Yogyakarta",
+      "instance_data_dir": "/content/data/Candi Kalasan - Yogyakarta",
+      "class_data_dir": "/content/data_class/Candi Kalasan - Yogyakarta"
+    },
+    {
+      "instance_prompt": "photo of Candi Mendut in Magelang",
+      "class_prompt": "photo of a Candi Mendut in Magelang",
+      "instance_data_dir": "/content/data/Candi Mendut - Magelang",
+      "class_data_dir": "/content/data_class/Candi Mendut - Magelang"
+    },
+    {
+      "instance_prompt": "photo of Candi Prambanan in Yogyakarta",
+      "class_prompt": "photo of a Candi Prambanan in Yogyakarta",
+      "instance_data_dir": "/content/data/Candi Prambanan - Yogyakarta",
+      "class_data_dir": "/content/data_class/Candi Prambanan - Yogyakarta"
+    },
+    {
+      "instance_prompt": "photo of Candi Sewu in Yogyakarta",
+      "class_prompt": "photo of a Candi Sewu in Yogyakarta",
+      "instance_data_dir": "/content/data/Candi Sewu - Yogyakarta",
+      "class_data_dir": "/content/data_class/Candi Sewu - Yogyakarta"
+    },
+    {
+      "instance_prompt": "photo of Danau Toba in Sumatera Utara",
+      "class_prompt": "photo of a Danau Toba in Sumatera Utara",
+      "instance_data_dir": "/content/data/Danau Toba - Sumatera Utara",
+      "class_data_dir": "/content/data_class/Danau Toba - Sumatera Utara"
+    },
+    {
+      "instance_prompt": "photo of Gedung Sate in Bandung",
+      "class_prompt": "photo of a Gedung Sate in Bandung",
+      "instance_data_dir": "/content/data/Gedung Sate - Bandung",
+      "class_data_dir": "/content/data_class/Gedung Sate - Bandung"
+    },
+    {
+      "instance_prompt": "photo of Istana Bogor in Bogor",
+      "class_prompt": "photo of a Istana Bogor in Bogor",
+      "instance_data_dir": "/content/data/Istana Bogor - Bogor",
+      "class_data_dir": "/content/data_class/Istana Bogor - Bogor"
+    },
+    {
+      "instance_prompt": "photo of Istana Merdeka in Jakarta",
+      "class_prompt": "photo of a Istana Merdeka in Jakarta",
+      "instance_data_dir": "/content/data/Istana Merdeka - Jakarta",
+      "class_data_dir": "/content/data_class/Istana Merdeka - Jakarta"
+    },
+    {
+      "instance_prompt": "photo of jokowi",
+      "class_prompt": "photo of a jokowi",
+      "instance_data_dir": "/content/data/jokowi",
+      "class_data_dir": "/content/data_class/jokowi"
+    },
+    {
+      "instance_prompt": "photo of Keraton Kasepuhan in Cirebon",
+      "class_prompt": "photo of a Keraton Kasepuhan in Cirebon",
+      "instance_data_dir": "/content/data/Keraton Kasepuhan - Cirebon",
+      "class_data_dir": "/content/data_class/Keraton Kasepuhan - Cirebon"
+    },
+    {
+      "instance_prompt": "photo of Kota Tua Jakarta in Jakarta",
+      "class_prompt": "photo of a Kota Tua Jakarta in Jakarta",
+      "instance_data_dir": "/content/data/Kota Tua Jakarta - Jakarta",
+      "class_data_dir": "/content/data_class/Kota Tua Jakarta - Jakarta"
+    },
+    {
+      "instance_prompt": "photo of Masjid Istiqlal in Jakarta",
+      "class_prompt": "photo of a Masjid Istiqlal in Jakarta",
+      "instance_data_dir": "/content/data/Masjid Istiqlal - Jakarta",
+      "class_data_dir": "/content/data_class/Masjid Istiqlal - Jakarta"
+    },
+    {
+      "instance_prompt": "photo of Monas in Jakarta",
+      "class_prompt": "photo of a Monas in Jakarta",
+      "instance_data_dir": "/content/data/Monas - Jakarta",
+      "class_data_dir": "/content/data_class/Monas - Jakarta"
+    },
+    {
+      "instance_prompt": "photo of Monumen Pancasila Sakti in Lubang Buaya",
+      "class_prompt": "photo of a Monumen Pancasila Sakti in Lubang Buaya",
+      "instance_data_dir": "/content/data/Monumen Pancasila Sakti - Lubang Buaya",
+      "class_data_dir": "/content/data_class/Monumen Pancasila Sakti - Lubang Buaya"
+    },
+    {
+      "instance_prompt": "photo of Monumen Simpang Lima Gumul in Kediri",
+      "class_prompt": "photo of a Monumen Simpang Lima Gumul in Kediri",
+      "instance_data_dir": "/content/data/Monumen Simpang Lima Gumul - Kediri",
+      "class_data_dir": "/content/data_class/Monumen Simpang Lima Gumul - Kediri"
+    },
+    {
+      "instance_prompt": "photo of Quraish Shihab, ahli tafsir Al-Quran",
+      "class_prompt": "photo of a Quraish Shihab, ahli tafsir Al-Quran",
+      "instance_data_dir": "/content/data/Quraish Shihab, ahli tafsir Al-Quran",
+      "class_data_dir": "/content/data_class/Quraish Shihab, ahli tafsir Al-Quran"
+    },
+    {
+      "instance_prompt": "photo of Rumah Gadang in Sumatera Barat",
+      "class_prompt": "photo of a Rumah Gadang in Sumatera Barat",
+      "instance_data_dir": "/content/data/Rumah Gadang - Sumatera Barat",
+      "class_data_dir": "/content/data_class/Rumah Gadang - Sumatera Barat"
+    },
+    {
+      "instance_prompt": "photo of Tugu Jogja in Yogyakarta",
+      "class_prompt": "photo of a Tugu Jogja in Yogyakarta",
+      "instance_data_dir": "/content/data/Tugu Jogja - Yogyakarta",
+      "class_data_dir": "/content/data_class/Tugu Jogja - Yogyakarta"
+    },
+    {
+      "instance_prompt": "photo of Tugu Monumen Simpang Lima in Semarang",
+      "class_prompt": "photo of a Tugu Monumen Simpang Lima in Semarang",
+      "instance_data_dir": "/content/data/Tugu Monumen Simpang Lima - Semarang",
+      "class_data_dir": "/content/data_class/Tugu Monumen Simpang Lima - Semarang"
+    },
+    {
+      "instance_prompt": "photo of Tugu Pahlawan in Surabaya",
+      "class_prompt": "photo of a Tugu Pahlawan in Surabaya",
+      "instance_data_dir": "/content/data/Tugu Pahlawan - Surabaya",
+      "class_data_dir": "/content/data_class/Tugu Pahlawan - Surabaya"
+    }
+  ],
+  "read_prompts_from_txts": false
+}

feature_extractor/preprocessor_config.json

@@ -0,0 +1,28 @@
+{
+  "crop_size": {
+    "height": 224,
+    "width": 224
+  },
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "feature_extractor_type": "CLIPFeatureExtractor",
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "CLIPFeatureExtractor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 224
+  }
+}

model.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06e0a622ca50b7eec606e2747581ca2ceb9a8b672b73cc7955377f81f19b4ac0
+size 2132791380

model_index.json

@@ -0,0 +1,33 @@
+{
+  "_class_name": "StableDiffusionPipeline",
+  "_diffusers_version": "0.15.0.dev0",
+  "feature_extractor": [
+    "transformers",
+    "CLIPFeatureExtractor"
+  ],
+  "requires_safety_checker": true,
+  "safety_checker": [
+    null,
+    null
+  ],
+  "scheduler": [
+    "diffusers",
+    "PNDMScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "CLIPTextModel"
+  ],
+  "tokenizer": [
+    "transformers",
+    "CLIPTokenizer"
+  ],
+  "unet": [
+    "diffusers",
+    "UNet2DConditionModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ]
+}

scheduler/scheduler_config.json

@@ -0,0 +1,18 @@
+{
+  "_class_name": "DDIMScheduler",
+  "_diffusers_version": "0.15.0.dev0",
+  "beta_end": 0.012,
+  "beta_schedule": "scaled_linear",
+  "beta_start": 0.00085,
+  "clip_sample": false,
+  "clip_sample_range": 1.0,
+  "dynamic_thresholding_ratio": 0.995,
+  "num_train_timesteps": 1000,
+  "prediction_type": "epsilon",
+  "sample_max_value": 1.0,
+  "set_alpha_to_one": false,
+  "skip_prk_steps": true,
+  "steps_offset": 1,
+  "thresholding": false,
+  "trained_betas": null
+}

text_encoder/config.json

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "runwayml/stable-diffusion-v1-5",
+  "architectures": [
+    "CLIPTextModel"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "dropout": 0.0,
+  "eos_token_id": 2,
+  "hidden_act": "quick_gelu",
+  "hidden_size": 768,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 77,
+  "model_type": "clip_text_model",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "projection_dim": 768,
+  "torch_dtype": "float32",
+  "transformers_version": "4.29.2",
+  "vocab_size": 49408
+}

text_encoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:036a18b3efed803068d5f8c2ac0e42effd9886a8b7bac6c795dc560528c5bcf5
+size 492309793

tokenizer/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer/tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "clean_up_tokenization_spaces": true,
+  "do_lower_case": true,
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "errors": "replace",
+  "model_max_length": 77,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "CLIPTokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

unet/config.json

@@ -0,0 +1,52 @@
+{
+  "_class_name": "UNet2DConditionModel",
+  "_diffusers_version": "0.15.0.dev0",
+  "_name_or_path": "runwayml/stable-diffusion-v1-5",
+  "act_fn": "silu",
+  "attention_head_dim": 8,
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "class_embeddings_concat": false,
+  "conv_in_kernel": 3,
+  "conv_out_kernel": 3,
+  "cross_attention_dim": 768,
+  "down_block_types": [
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "dual_cross_attention": false,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 4,
+  "layers_per_block": 2,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock2DCrossAttn",
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 4,
+  "projection_class_embeddings_input_dim": null,
+  "resnet_time_scale_shift": "default",
+  "sample_size": 64,
+  "time_cond_proj_dim": null,
+  "time_embedding_type": "positional",
+  "timestep_post_act": null,
+  "up_block_types": [
+    "UpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D"
+  ],
+  "upcast_attention": false,
+  "use_linear_projection": false
+}

unet/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f6c40ef51055f5a61a84d0e5159a6208453d21b5b4d084b92c9bb18ca591693
+size 3438375973

vae/config.json

@@ -0,0 +1,31 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.15.0.dev0",
+  "_name_or_path": "stabilityai/sd-vae-ft-mse",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "in_channels": 3,
+  "latent_channels": 4,
+  "layers_per_block": 2,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 256,
+  "scaling_factor": 0.18215,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ]
+}

vae/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36bb8e1b54aba3a0914eb35fba13dcb107e9f18d379d1df2158732cd4bf56a94
+size 334711857