Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,25 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+samples/1754680954383__000000000_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754682897160__000000250_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754684855471__000000500_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754686558622__000000750_0.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754686810772__000000750_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754688766311__000001000_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754690468085__000001250_0.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754690720285__000001250_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754692422398__000001500_0.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754692674400__000001500_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754694377605__000001750_0.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754694629787__000001750_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754696333775__000002000_0.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754696585958__000002000_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754698293007__000002250_0.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754698545223__000002250_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754700253716__000002500_0.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754700505906__000002500_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754702213594__000002750_0.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754702465712__000002750_1.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754704167607__000003000_0.jpg filter=lfs diff=lfs merge=lfs -text
+samples/1754704419868__000003000_1.jpg filter=lfs diff=lfs merge=lfs -text

.job_config.json

@@ -0,0 +1,120 @@
+{
+  "job": "extension",
+  "config": {
+    "name": "skilgrimr_v1",
+    "process": [
+      {
+        "type": "ui_trainer",
+        "training_folder": "/workspace/ai-toolkit/output",
+        "sqlite_db_path": "/workspace/ai-toolkit/aitk_db.db",
+        "device": "cuda",
+        "trigger_word": "skilgrimr",
+        "performance_log_every": 10,
+        "network": {
+          "type": "lora",
+          "linear": 32,
+          "linear_alpha": 32,
+          "conv": 16,
+          "conv_alpha": 16,
+          "lokr_full_rank": true,
+          "lokr_factor": -1,
+          "network_kwargs": {
+            "ignore_if_contains": []
+          }
+        },
+        "save": {
+          "dtype": "bf16",
+          "save_every": 250,
+          "max_step_saves_to_keep": 4,
+          "save_format": "diffusers",
+          "push_to_hub": false
+        },
+        "datasets": [
+          {
+            "folder_path": "/workspace/ai-toolkit/datasets/skilgrimr",
+            "control_path": null,
+            "mask_path": null,
+            "mask_min_value": 0.1,
+            "default_caption": "",
+            "caption_ext": "txt",
+            "caption_dropout_rate": 0.05,
+            "cache_latents_to_disk": false,
+            "is_reg": false,
+            "network_weight": 1,
+            "resolution": [
+              1024
+            ],
+            "controls": [],
+            "shrink_video_to_frames": true,
+            "num_frames": 1,
+            "do_i2v": true
+          }
+        ],
+        "train": {
+          "batch_size": 1,
+          "bypass_guidance_embedding": false,
+          "steps": 3000,
+          "gradient_accumulation": 1,
+          "train_unet": true,
+          "train_text_encoder": false,
+          "gradient_checkpointing": true,
+          "noise_scheduler": "flowmatch",
+          "optimizer": "adamw8bit",
+          "timestep_type": "weighted",
+          "content_or_style": "balanced",
+          "optimizer_params": {
+            "weight_decay": 0.0001
+          },
+          "unload_text_encoder": false,
+          "cache_text_embeddings": false,
+          "lr": 0.0001,
+          "ema_config": {
+            "use_ema": false,
+            "ema_decay": 0.99
+          },
+          "skip_first_sample": false,
+          "disable_sampling": false,
+          "dtype": "bf16",
+          "diff_output_preservation": false,
+          "diff_output_preservation_multiplier": 1,
+          "diff_output_preservation_class": "person"
+        },
+        "model": {
+          "name_or_path": "Qwen/Qwen-Image",
+          "quantize": true,
+          "qtype": "qfloat8",
+          "quantize_te": true,
+          "qtype_te": "qfloat8",
+          "arch": "qwen_image",
+          "low_vram": true,
+          "model_kwargs": {}
+        },
+        "sample": {
+          "sampler": "flowmatch",
+          "sample_every": 250,
+          "width": 1328,
+          "height": 1328,
+          "samples": [
+            {
+              "prompt": "portrait of skilgrimr, neutral gray seamless backdrop, 85mm lens look, soft key + gentle fill, mouth closed, slight smile, sharp eyes, realistic skin texture, minimal shadows, photographic"
+            },
+            {
+              "prompt": "rain-soaked night street, close-up portrait of skilgrimr, neon reflections and teal–magenta bokeh, shallow depth of field, wet hair detail on skilgrimr, cinematic cyberpunk mood, specular highlights on skilgrimr’s skin, 35mm lens look, no hood, no high collar          Ask ChatGPT"
+            }
+          ],
+          "neg": "",
+          "seed": 42,
+          "walk_seed": true,
+          "guidance_scale": 4,
+          "sample_steps": 50,
+          "num_frames": 1,
+          "fps": 1
+        }
+      }
+    ]
+  },
+  "meta": {
+    "name": "[name]",
+    "version": "1.0"
+  }
+}

config.yaml

@@ -0,0 +1,101 @@
+job: extension
+config:
+  name: skilgrimr_v1
+  process:
+  - type: ui_trainer
+    training_folder: /workspace/ai-toolkit/output
+    sqlite_db_path: /workspace/ai-toolkit/aitk_db.db
+    device: cuda
+    trigger_word: skilgrimr
+    performance_log_every: 10
+    network:
+      type: lora
+      linear: 32
+      linear_alpha: 32
+      conv: 16
+      conv_alpha: 16
+      lokr_full_rank: true
+      lokr_factor: -1
+      network_kwargs:
+        ignore_if_contains: []
+    save:
+      dtype: bf16
+      save_every: 250
+      max_step_saves_to_keep: 4
+      save_format: diffusers
+      push_to_hub: false
+    datasets:
+    - folder_path: /workspace/ai-toolkit/datasets/skilgrimr
+      control_path: null
+      mask_path: null
+      mask_min_value: 0.1
+      default_caption: ''
+      caption_ext: txt
+      caption_dropout_rate: 0.05
+      cache_latents_to_disk: false
+      is_reg: false
+      network_weight: 1
+      resolution:
+      - 1024
+      controls: []
+      shrink_video_to_frames: true
+      num_frames: 1
+      do_i2v: true
+    train:
+      batch_size: 1
+      bypass_guidance_embedding: false
+      steps: 3000
+      gradient_accumulation: 1
+      train_unet: true
+      train_text_encoder: false
+      gradient_checkpointing: true
+      noise_scheduler: flowmatch
+      optimizer: adamw8bit
+      timestep_type: weighted
+      content_or_style: balanced
+      optimizer_params:
+        weight_decay: 0.0001
+      unload_text_encoder: false
+      cache_text_embeddings: false
+      lr: 0.0001
+      ema_config:
+        use_ema: false
+        ema_decay: 0.99
+      skip_first_sample: false
+      disable_sampling: false
+      dtype: bf16
+      diff_output_preservation: false
+      diff_output_preservation_multiplier: 1
+      diff_output_preservation_class: person
+    model:
+      name_or_path: Qwen/Qwen-Image
+      quantize: true
+      qtype: qfloat8
+      quantize_te: true
+      qtype_te: qfloat8
+      arch: qwen_image
+      low_vram: true
+      model_kwargs: {}
+    sample:
+      sampler: flowmatch
+      sample_every: 250
+      width: 1328
+      height: 1328
+      samples:
+      - prompt: portrait of skilgrimr, neutral gray seamless backdrop, 85mm lens look,
+          soft key + gentle fill, mouth closed, slight smile, sharp eyes, realistic
+          skin texture, minimal shadows, photographic
+      - prompt: "rain-soaked night street, close-up portrait of skilgrimr, neon reflections\
+          \ and teal\u2013magenta bokeh, shallow depth of field, wet hair detail on\
+          \ skilgrimr, cinematic cyberpunk mood, specular highlights on skilgrimr\u2019\
+          s skin, 35mm lens look, no hood, no high collar          Ask ChatGPT"
+      neg: ''
+      seed: 42
+      walk_seed: true
+      guidance_scale: 4
+      sample_steps: 50
+      num_frames: 1
+      fps: 1
+meta:
+  name: skilgrimr_v1
+  version: '1.0'

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9a45b13aa0751258bf376558b97f34be48fb1103fbdc0bac72fc6e277ad44fd
+size 598613389

skilgrimr_v1.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91e3fb4512700a9a68a407236bd1c780314896141470f1b85c49648cebe15c4d
+size 590058880

skilgrimr_v1_000002000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ea4e8046cc008813acdcec206d526b44ccea365b865d12dcabdce4d135c3d38
+size 590058880

skilgrimr_v1_000002250.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b7fc2e10e20799ca43fcc42f219a8b48cd33c3420ef7f15052de326d5b5904b
+size 590058880

skilgrimr_v1_000002500.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35023fe9cc594f6ab0b17bde8b58040b6d68aea1d6b7e7f9765ebae6fb79c656
+size 590058880

skilgrimr_v1_000002750.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ca9803cb8061cc42cba8d10fb682d3cde9778ae5bb25a7832720fb790c496ac
+size 590058880