Initial commit
Browse files- README.md +52 -1
- model_index.json +12 -0
- scheduler/scheduler_config.json +11 -0
- unet/config.json +42 -0
- unet/diffusion_pytorch_model.bin +3 -0
README.md
CHANGED
|
@@ -1,3 +1,54 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
language: en
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
library_name: diffusers
|
| 5 |
+
tags: []
|
| 6 |
+
datasets: huggan/pokemon
|
| 7 |
+
metrics: []
|
| 8 |
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the training script had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# ddpm-ema-pokemon-64
|
| 14 |
+
|
| 15 |
+
## Model description
|
| 16 |
+
|
| 17 |
+
This diffusion model is trained with the [🤗 Diffusers](https://github.com/huggingface/diffusers) library
|
| 18 |
+
on the `huggan/pokemon` dataset.
|
| 19 |
+
|
| 20 |
+
## Intended uses & limitations
|
| 21 |
+
|
| 22 |
+
#### How to use
|
| 23 |
+
|
| 24 |
+
```python
|
| 25 |
+
# TODO: add an example code snippet for running this diffusion pipeline
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
#### Limitations and bias
|
| 29 |
+
|
| 30 |
+
[TODO: provide examples of latent issues and potential remediations]
|
| 31 |
+
|
| 32 |
+
## Training data
|
| 33 |
+
|
| 34 |
+
[TODO: describe the data used to train the model]
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 0.0001
|
| 40 |
+
- train_batch_size: 256
|
| 41 |
+
- eval_batch_size: 128
|
| 42 |
+
- gradient_accumulation_steps: 1
|
| 43 |
+
- optimizer: AdamW with betas=(0.95, 0.999), weight_decay=1e-06 and epsilon=1e-08
|
| 44 |
+
- lr_scheduler: None
|
| 45 |
+
- lr_warmup_steps: 500
|
| 46 |
+
- ema_inv_gamma: 1.0
|
| 47 |
+
- ema_inv_gamma: 0.75
|
| 48 |
+
- ema_inv_gamma: 0.9999
|
| 49 |
+
- mixed_precision: fp16
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
📈 [TensorBoard logs](https://huggingface.co/mrm8488/ddpm-ema-pokemon-64/tensorboard?#scalars)
|
| 54 |
+
|
model_index.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "DDPMPipeline",
|
| 3 |
+
"_diffusers_version": "0.1.2",
|
| 4 |
+
"scheduler": [
|
| 5 |
+
"diffusers",
|
| 6 |
+
"DDPMScheduler"
|
| 7 |
+
],
|
| 8 |
+
"unet": [
|
| 9 |
+
"diffusers",
|
| 10 |
+
"UNet2DModel"
|
| 11 |
+
]
|
| 12 |
+
}
|
scheduler/scheduler_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "DDPMScheduler",
|
| 3 |
+
"_diffusers_version": "0.1.2",
|
| 4 |
+
"beta_end": 0.02,
|
| 5 |
+
"beta_schedule": "linear",
|
| 6 |
+
"beta_start": 0.0001,
|
| 7 |
+
"clip_sample": true,
|
| 8 |
+
"num_train_timesteps": 1000,
|
| 9 |
+
"trained_betas": null,
|
| 10 |
+
"variance_type": "fixed_small"
|
| 11 |
+
}
|
unet/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "UNet2DModel",
|
| 3 |
+
"_diffusers_version": "0.1.2",
|
| 4 |
+
"act_fn": "silu",
|
| 5 |
+
"attention_head_dim": 8,
|
| 6 |
+
"block_out_channels": [
|
| 7 |
+
128,
|
| 8 |
+
128,
|
| 9 |
+
256,
|
| 10 |
+
256,
|
| 11 |
+
512,
|
| 12 |
+
512
|
| 13 |
+
],
|
| 14 |
+
"center_input_sample": false,
|
| 15 |
+
"down_block_types": [
|
| 16 |
+
"DownBlock2D",
|
| 17 |
+
"DownBlock2D",
|
| 18 |
+
"DownBlock2D",
|
| 19 |
+
"DownBlock2D",
|
| 20 |
+
"AttnDownBlock2D",
|
| 21 |
+
"DownBlock2D"
|
| 22 |
+
],
|
| 23 |
+
"downsample_padding": 1,
|
| 24 |
+
"flip_sin_to_cos": true,
|
| 25 |
+
"freq_shift": 0,
|
| 26 |
+
"in_channels": 3,
|
| 27 |
+
"layers_per_block": 2,
|
| 28 |
+
"mid_block_scale_factor": 1,
|
| 29 |
+
"norm_eps": 1e-05,
|
| 30 |
+
"norm_num_groups": 32,
|
| 31 |
+
"out_channels": 3,
|
| 32 |
+
"sample_size": 64,
|
| 33 |
+
"time_embedding_type": "positional",
|
| 34 |
+
"up_block_types": [
|
| 35 |
+
"UpBlock2D",
|
| 36 |
+
"AttnUpBlock2D",
|
| 37 |
+
"UpBlock2D",
|
| 38 |
+
"UpBlock2D",
|
| 39 |
+
"UpBlock2D",
|
| 40 |
+
"UpBlock2D"
|
| 41 |
+
]
|
| 42 |
+
}
|
unet/diffusion_pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:abd90cf7254431ee4e83965c968022d1c0587ee34e8430fda58290c5fac4b449
|
| 3 |
+
size 454859389
|