Hugging Face's logo

xyfJASON
/
PartialConv-Inpainting-pytorch

TensorBoard
Model card Files Metrics Community
xyfJASON commited on
Commit
2d3f7b2
·
verified ·
1 Parent(s): 9622996

Upload celebahq-256-finetune checkpoint and training logs

Browse files
Files changed (6) hide show
  1. celebahq-256-finetune/ckpt/best/model.pt +3 -0
  2. celebahq-256-finetune/ckpt/step069999/model.pt +3 -0
  3. celebahq-256-finetune/config-2023-02-08-02-46-53.yaml +55 -0
  4. celebahq-256-finetune/output-2023-02-08-02-46-53.log +0 -0
  5. celebahq-256-finetune/samples.zip +3 -0
  6. celebahq-256-finetune/tensorboard/events.out.tfevents.1675824416.jason-system.7508.0 +3 -0
celebahq-256-finetune/ckpt/best/model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cfdccf17ba8b9da53eccf4cbbc3c14712c7014c657733248f3896315d789d87
3
+ size 262939477
celebahq-256-finetune/ckpt/step069999/model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1bc59d0cdb64e8f40c5758d411f39ac3dd45df7f74721ffee19c515b1e01eb0
3
+ size 262939477
celebahq-256-finetune/config-2023-02-08-02-46-53.yaml ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ DATA:
2
+ DATAROOT: /data/CelebA-HQ/
3
+ IMG_SIZE: 256
4
+ NAME: CelebA-HQ
5
+ DATALOADER:
6
+ BATCH_SIZE: 12
7
+ MICRO_BATCH: 0
8
+ NUM_WORKERS: 4
9
+ PIN_MEMORY: true
10
+ PREFETCH_FACTOR: 2
11
+ EVALUATE:
12
+ N_EVAL: 0
13
+ MASK:
14
+ BRUSH_LENGTH_RATIO:
15
+ - 0.1
16
+ - 0.25
17
+ BRUSH_NUM: &id001
18
+ - 0
19
+ - 0
20
+ BRUSH_TURNS:
21
+ - 4
22
+ - 18
23
+ BRUSH_WIDTH_RATIO:
24
+ - 0.02
25
+ - 0.1
26
+ MASK_DIR: /data/NVIDIAIrregularMaskDataset/train/
27
+ RECT_LENGTH_RATIO:
28
+ - 0.2
29
+ - 0.8
30
+ RECT_NUM: *id001
31
+ MODEL:
32
+ N_LAYERS: 8
33
+ WEIGHTS: ./runs/celebahq-256/ckpt/step199999.pt
34
+ SAMPLE:
35
+ N_SAMPLES: 100
36
+ RANDOM: false
37
+ SAVE_DIR: ./samples/
38
+ SEED: 1234
39
+ TRAIN:
40
+ COEF_HOLE: 6.0
41
+ COEF_PERCEPTUAL: 0.05
42
+ COEF_STYLE: 120
43
+ COEF_TV: 0.1
44
+ COEF_VALID: 1.0
45
+ EVAL_FREQ: 1000
46
+ OPTIM:
47
+ BETAS:
48
+ - 0.5
49
+ - 0.999
50
+ LR: 5.0e-05
51
+ PRINT_FREQ: 100
52
+ RESUME: null
53
+ SAMPLE_FREQ: 1000
54
+ SAVE_FREQ: 5000
55
+ TRAIN_STEPS: 70000
celebahq-256-finetune/output-2023-02-08-02-46-53.log ADDED
The diff for this file is too large to render. See raw diff
 
celebahq-256-finetune/samples.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d55dcfaf84aabd2a0ed251c6a9390116851ab93cf12ef1f712bf74a41956d55
3
+ size 114824140
celebahq-256-finetune/tensorboard/events.out.tfevents.1675824416.jason-system.7508.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b60f396c4910e9be3731e3c8c703a27571ac1906cb49dd1d4001f0a828743318
3
+ size 19597572