0.5.3

block in a section that links to available pre-trained models for those who are interested

README.md

@@ -12,7 +12,7 @@ This model is SOTA for text-to-image for now.
 
 Please join <a href="https://discord.gg/xBPBXfcFHd"><img alt="Join us on Discord" src="https://img.shields.io/discord/823813159592001537?color=5865F2&logo=discord&logoColor=white"></a> if you are interested in helping out with the replication with the <a href="https://laion.ai/">LAION</a> community | <a href="https://www.youtube.com/watch?v=AIOE1l1W0Tw">Yannic Interview</a>
 
-There was enough interest for a <a href="https://github.com/lucidrains/dalle2-jax">Jax version</a>. I will also eventually extend this to <a href="https://github.com/lucidrains/dalle2-video">text to video</a>, once the repository is in a good place.
+As of 5/23/22, it is no longer SOTA. SOTA will be <a href="https://github.com/lucidrains/imagen-pytorch">here</a>. Jax versions as well as text-to-video project will be shifted towards the Imagen architecture, as it is way simpler.
 
 ## Status
 
@@ -24,6 +24,11 @@ There was enough interest for a <a href="https://github.com/lucidrains/dalle2-ja
 
 *ongoing at 21k steps*
 
+## Pre-Trained Models
+- LAION is training prior models. Checkpoints are available on <a href="https://huggingface.co/zenglishuci/conditioned-prior">🤗huggingface</a> and the training statistics are available on <a href="https://wandb.ai/nousr_laion/conditioned-prior/reports/LAION-DALLE2-PyTorch-Prior--VmlldzoyMDI2OTIx">🐝WANDB</a>.
+- Decoder - <a href="https://wandb.ai/veldrovive/dalle2_train_decoder/runs/jkrtg0so?workspace=user-veldrovive">In-progress test run</a> 🚧
+- DALL-E 2 🚧
+
 ## Install
 
 ```bash
@@ -1077,6 +1082,9 @@ This library would not have gotten to this working state without the help of
 - [x] cross embed layers for downsampling, as an option
 - [x] use an experimental tracker agnostic setup, as done <a href="https://github.com/lucidrains/tf-bind-transformer#simple-trainer-class-for-fine-tuning">here</a>
 - [x] use pydantic for config drive training
+- [x] for both diffusion prior and decoder, all exponential moving averaged models needs to be saved and restored as well (as well as the step number)
+- [x] offer save / load methods on the trainer classes to automatically take care of state dicts for scalers / optimizers / saving versions and checking for breaking changes
+- [x] allow for creation of diffusion prior model off pydantic config classes - consider the same for tracker configs
 - [ ] become an expert with unets, cleanup unet code, make it fully configurable, port all learnings over to https://github.com/lucidrains/x-unet (test out unet² in ddpm repo) - consider https://github.com/lucidrains/uformer-pytorch attention-based unet
 - [ ] transcribe code to Jax, which lowers the activation energy for distributed training, given access to TPUs
 - [ ] train on a toy task, offer in colab
@@ -1086,12 +1094,9 @@ This library would not have gotten to this working state without the help of
 - [ ] test out grid attention in cascading ddpm locally, decide whether to keep or remove
 - [ ] interface out the vqgan-vae so a pretrained one can be pulled off the shelf to validate latent diffusion + DALL-E2
 - [ ] make sure FILIP works with DALL-E2 from x-clip https://arxiv.org/abs/2111.07783
-- [ ] offer save / load methods on the trainer classes to automatically take care of state dicts for scalers / optimizers / saving versions and checking for breaking changes
 - [ ] bring in skip-layer excitatons (from lightweight gan paper) to see if it helps for either decoder of unet or vqgan-vae training
 - [ ] decoder needs one day worth of refactor for tech debt
 - [ ] allow for unet to be able to condition non-cross attention style as well
-- [ ] for all model classes with hyperparameters that changes the network architecture, make it requirement that they must expose a config property, and write a simple function that asserts that it restores the object correctly
-- [ ] for both diffusion prior and decoder, all exponential moving averaged models needs to be saved and restored as well (as well as the step number)
 - [ ] read the paper, figure it out, and build it https://github.com/lucidrains/DALLE2-pytorch/issues/89
 
 ## Citations
@@ -1190,4 +1195,12 @@ This library would not have gotten to this working state without the help of
 }
 ```
 
+```bibtex
+@misc{Saharia2022,
+    title   = {Imagen: unprecedented photorealism × deep level of language understanding},
+    author  = {Chitwan Saharia*, William Chan*, Saurabh Saxena†, Lala Li†, Jay Whang†, Emily Denton, Seyed Kamyar Seyed Ghasemipour, Burcu Karagol Ayan, S. Sara Mahdavi, Rapha Gontijo Lopes, Tim Salimans, Jonathan Ho†, David Fleet†, Mohammad Norouzi*},
+    year    = {2022}
+}
+```
+
 *Creating noise from data is easy; creating data from noise is generative modeling.* - <a href="https://arxiv.org/abs/2011.13456">Yang Song's paper</a>

configs/README.md

@@ -6,9 +6,10 @@ For more complex configuration, we provide the option of using a configuration f
 
 The decoder trainer has 7 main configuration options. A full example of their use can be found in the [example decoder configuration](train_decoder_config.example.json).
 
-**<ins>Unets</ins>:**
+**<ins>Unet</ins>:**
+
+This is a single unet config, which belongs as an array nested under the decoder config as a list of `unets`
 
-Each member of this array defines a single unet that will be added to the decoder.
 | Option | Required | Default | Description |
 | ------ | -------- | ------- | ----------- |
 | `dim`  | Yes      | N/A     | The starting channels of the unet. |
@@ -22,6 +23,7 @@ Any parameter from the `Unet` constructor can also be given here.
 Defines the configuration options for the decoder model. The unets defined above will automatically be inserted.
 | Option | Required | Default | Description |
 | ------ | -------- | ------- | ----------- |
+| `unets` | Yes | N/A | A list of unets, using the configuration above |
 | `image_sizes` | Yes | N/A | The resolution of the image after each upsampling step. The length of this array should be the number of unets defined. |
 | `image_size` | Yes | N/A | Not used. Can be any number. |
 | `timesteps` | No | `1000` | The number of diffusion timesteps used for generation. |

configs/train_decoder_config.example.json

@@ -1,16 +1,16 @@
 {
-    "unets": [
-        {
-            "dim": 128,
-            "image_embed_dim": 768,
-            "cond_dim": 64,
-            "channels": 3,
-            "dim_mults": [1, 2, 4, 8],
-            "attn_dim_head": 32,
-            "attn_heads": 16
-        }
-    ],
     "decoder": {
+        "unets": [
+            {
+                "dim": 128,
+                "image_embed_dim": 768,
+                "cond_dim": 64,
+                "channels": 3,
+                "dim_mults": [1, 2, 4, 8],
+                "attn_dim_head": 32,
+                "attn_heads": 16
+            }
+        ],
         "image_sizes": [64],
         "channels": 3,
         "timesteps": 1000,

configs/train_prior_config.example.json

@@ -0,0 +1,70 @@
+{
+    "prior": {
+        "clip": {
+            "make": "x-clip",
+            "model": "ViT-L/14",
+            "base_model_kwargs": {
+                "dim_text": 768,
+                "dim_image": 768,
+                "dim_latent": 768
+            }
+        },
+        "net": {
+            "dim": 768,
+            "depth": 12,
+            "num_timesteps": 1000,
+            "num_time_embeds": 1,
+            "num_image_embeds": 1,
+            "num_text_embeds": 1,
+            "dim_head": 64,
+            "heads": 12,
+            "ff_mult": 4,
+            "norm_out": true,
+            "attn_dropout": 0.0,
+            "ff_dropout": 0.0,
+            "final_proj": true,
+            "normformer": true,
+            "rotary_emb": true
+        },
+        "image_embed_dim": 768,
+        "image_size": 224,
+        "image_channels": 3,
+        "timesteps": 1000,
+        "cond_drop_prob": 0.1,
+        "loss_type": "l2",
+        "predict_x_start": true,
+        "beta_schedule": "cosine",
+        "condition_on_text_encodings": true
+    },
+    "data": {
+        "image_url": "https://mystic.the-eye.eu/public/AI/cah/laion5b/embeddings/laion2B-en/img_emb/",
+        "text_url": "https://mystic.the-eye.eu/public/AI/cah/laion5b/embeddings/laion2B-en/text_emb/",
+        "meta_url": "https://mystic.the-eye.eu/public/AI/cah/laion5b/embeddings/laion2B-en/laion2B-en-metadata/",
+        "batch_size": 256,
+        "splits": {
+            "train": 0.9,
+            "val": 1e-7,
+            "test": 0.0999999
+        }
+    },
+    "train": {
+        "epochs": 1,
+        "lr": 1.1e-4,
+        "wd": 6.02e-2,
+        "max_grad_norm": 0.5,
+        "use_ema": true,
+        "amp": false,
+        "save_every": 10000
+    },
+    "load": {
+        "source": null,
+        "resume": false
+    },
+    "tracker": {
+        "tracker_type": "wandb",
+        "data_path": "./prior_checkpoints",
+        "wandb_entity": "laion",
+        "wandb_project": "diffusion-prior",
+        "verbose": true
+    }
+}

dalle2_pytorch/dalle2_pytorch.py

@@ -890,6 +890,8 @@ class DiffusionPrior(BaseGaussianDiffusion):
         )
 
         if exists(clip):
+            assert image_channels == clip.image_channels, f'channels of image ({image_channels}) should be equal to the channels that CLIP accepts ({clip.image_channels})'
+
             if isinstance(clip, CLIP):
                 clip = XClipAdapter(clip, **clip_adapter_overrides)
             elif isinstance(clip, CoCa):
@@ -1105,13 +1107,20 @@ class Block(nn.Module):
         groups = 8
     ):
         super().__init__()
-        self.block = nn.Sequential(
-            nn.Conv2d(dim, dim_out, 3, padding = 1),
-            nn.GroupNorm(groups, dim_out),
-            nn.SiLU()
-        )
-    def forward(self, x):
-        return self.block(x)
+        self.project = nn.Conv2d(dim, dim_out, 3, padding = 1)
+        self.norm = nn.GroupNorm(groups, dim_out)
+        self.act = nn.SiLU()
+
+    def forward(self, x, scale_shift = None):
+        x = self.project(x)
+        x = self.norm(x)
+
+        if exists(scale_shift):
+            scale, shift = scale_shift
+            x = x * (scale + 1) + shift
+
+        x = self.act(x)
+        return x
 
 class ResnetBlock(nn.Module):
     def __init__(
@@ -1130,7 +1139,7 @@ class ResnetBlock(nn.Module):
         if exists(time_cond_dim):
             self.time_mlp = nn.Sequential(
                 nn.SiLU(),
-                nn.Linear(time_cond_dim, dim_out)
+                nn.Linear(time_cond_dim, dim_out * 2)
             )
 
         self.cross_attn = None
@@ -1150,11 +1159,14 @@ class ResnetBlock(nn.Module):
         self.res_conv = nn.Conv2d(dim, dim_out, 1) if dim != dim_out else nn.Identity()
 
     def forward(self, x, cond = None, time_emb = None):
-        h = self.block1(x)
 
+        scale_shift = None
         if exists(self.time_mlp) and exists(time_emb):
             time_emb = self.time_mlp(time_emb)
-            h = rearrange(time_emb, 'b c -> b c 1 1') + h
+            time_emb = rearrange(time_emb, 'b c -> b c 1 1')
+            scale_shift = time_emb.chunk(2, dim = 1)
+
+        h = self.block1(x, scale_shift = scale_shift)
 
         if exists(self.cross_attn):
             assert exists(cond)
@@ -1702,6 +1714,8 @@ class Decoder(BaseGaussianDiffusion):
         vb_loss_weight = 0.001,
         unconditional = False,
         auto_normalize_img = True,                  # whether to take care of normalizing the image from [0, 1] to [-1, 1] and back automatically - you can turn this off if you want to pass in the [-1, 1] ranged image yourself from the dataloader
+        use_dynamic_thres = False,                  # from the Imagen paper
+        dynamic_thres_percentile = 0.9
     ):
         super().__init__(
             beta_schedule = beta_schedule,
@@ -1710,12 +1724,19 @@ class Decoder(BaseGaussianDiffusion):
         )
 
         self.unconditional = unconditional
-        assert not (condition_on_text_encodings and unconditional), 'unconditional decoder image generation cannot be set to True if conditioning on text is present'
 
-        assert self.unconditional or (exists(clip) ^ exists(image_size)), 'either CLIP is supplied, or you must give the image_size and channels (usually 3 for RGB)'
+        # text conditioning
+
+        assert not (condition_on_text_encodings and unconditional), 'unconditional decoder image generation cannot be set to True if conditioning on text is present'
+        self.condition_on_text_encodings = condition_on_text_encodings
+
+        # clip
 
         self.clip = None
         if exists(clip):
+            assert not unconditional, 'clip must not be given if doing unconditional image training'
+            assert channels == clip.image_channels, f'channels of image ({channels}) should be equal to the channels that CLIP accepts ({clip.image_channels})'
+
             if isinstance(clip, CLIP):
                 clip = XClipAdapter(clip, **clip_adapter_overrides)
             elif isinstance(clip, CoCa):
@@ -1725,13 +1746,20 @@ class Decoder(BaseGaussianDiffusion):
             assert isinstance(clip, BaseClipAdapter)
 
             self.clip = clip
-            self.clip_image_size = clip.image_size
-            self.channels = clip.image_channels
-        else:
-            self.clip_image_size = image_size
-            self.channels = channels
 
-        self.condition_on_text_encodings = condition_on_text_encodings
+        # determine image size, with image_size and image_sizes taking precedence
+
+        if exists(image_size) or exists(image_sizes):
+            assert exists(image_size) ^ exists(image_sizes), 'only one of image_size or image_sizes must be given'
+            image_size = default(image_size, lambda: image_sizes[-1])
+        elif exists(clip):
+            image_size = clip.image_size
+        else:
+            raise Error('either image_size, image_sizes, or clip must be given to decoder')
+
+        # channels
+
+        self.channels = channels
 
         # automatically take care of ensuring that first unet is unconditional
         # while the rest of the unets are conditioned on the low resolution image produced by previous unet
@@ -1773,7 +1801,7 @@ class Decoder(BaseGaussianDiffusion):
 
         # unet image sizes
 
-        image_sizes = default(image_sizes, (self.clip_image_size,))
+        image_sizes = default(image_sizes, (image_size,))
         image_sizes = tuple(sorted(set(image_sizes)))
 
         assert len(self.unets) == len(image_sizes), f'you did not supply the correct number of u-nets ({len(self.unets)}) for resolutions {image_sizes}'
@@ -1810,7 +1838,13 @@ class Decoder(BaseGaussianDiffusion):
         self.clip_denoised = clip_denoised
         self.clip_x_start = clip_x_start
 
+        # dynamic thresholding settings, if clipping denoised during sampling
+
+        self.use_dynamic_thres = use_dynamic_thres
+        self.dynamic_thres_percentile = dynamic_thres_percentile
+
         # normalize and unnormalize image functions
+
         self.normalize_img = normalize_neg_one_to_one if auto_normalize_img else identity
         self.unnormalize_img = unnormalize_zero_to_one if auto_normalize_img else identity
 
@@ -1851,7 +1885,21 @@ class Decoder(BaseGaussianDiffusion):
             x_recon = self.predict_start_from_noise(x, t = t, noise = pred)
 
         if clip_denoised:
-            x_recon.clamp_(-1., 1.)
+            # s is the threshold amount
+            # static thresholding would just be s = 1
+            s = 1.
+            if self.use_dynamic_thres:
+                s = torch.quantile(
+                    rearrange(x_recon, 'b ... -> b (...)').abs(),
+                    self.dynamic_thres_percentile,
+                    dim = -1
+                )
+
+                s.clamp_(min = 1.)
+                s = s.view(-1, *((1,) * (x_recon.ndim - 1)))
+
+            # clip by threshold, depending on whether static or dynamic
+            x_recon = x_recon.clamp(-s, s) / s
 
         model_mean, posterior_variance, posterior_log_variance = self.q_posterior(x_start=x_recon, x_t=x, t=t)
 

dalle2_pytorch/optimizer.py

@@ -12,6 +12,7 @@ def get_optimizer(
     betas = (0.9, 0.999),
     eps = 1e-8,
     filter_by_requires_grad = False,
+    group_wd_params = True,
     **kwargs
 ):
     if filter_by_requires_grad:
@@ -21,11 +22,13 @@ def get_optimizer(
         return Adam(params, lr = lr, betas = betas, eps = eps)
 
     params = set(params)
-    wd_params, no_wd_params = separate_weight_decayable_params(params)
 
-    param_groups = [
-        {'params': list(wd_params)},
-        {'params': list(no_wd_params), 'weight_decay': 0},
-    ]
+    if group_wd_params:
+        wd_params, no_wd_params = separate_weight_decayable_params(params)
 
-    return AdamW(param_groups, lr = lr, weight_decay = wd, betas = betas, eps = eps)
+        params = [
+            {'params': list(wd_params)},
+            {'params': list(no_wd_params), 'weight_decay': 0},
+        ]
+
+    return AdamW(params, lr = lr, weight_decay = wd, betas = betas, eps = eps)

dalle2_pytorch/train_configs.py

@@ -1,16 +1,152 @@
+import json
 from torchvision import transforms as T
 from pydantic import BaseModel, validator, root_validator
 from typing import List, Iterable, Optional, Union, Tuple, Dict, Any
 
+from x_clip import CLIP as XCLIP
+from coca_pytorch import CoCa
+
+from dalle2_pytorch.dalle2_pytorch import (
+    CoCaAdapter,
+    OpenAIClipAdapter,
+    Unet,
+    Decoder,
+    DiffusionPrior,
+    DiffusionPriorNetwork,
+    XClipAdapter,
+)
+
+# helper functions
+
 def exists(val):
     return val is not None
 
 def default(val, d):
     return val if exists(val) else d
 
+def ListOrTuple(inner_type):
+    return Union[List[inner_type], Tuple[inner_type]]
+
+# general pydantic classes
+
+class TrainSplitConfig(BaseModel):
+    train: float = 0.75
+    val: float = 0.15
+    test: float = 0.1
+
+    @root_validator
+    def validate_all(cls, fields):
+        actual_sum = sum([*fields.values()])
+        if actual_sum != 1.:
+            raise ValueError(f'{fields.keys()} must sum to 1.0. Found: {actual_sum}')
+        return fields
+
+class TrackerConfig(BaseModel):
+    tracker_type: str = 'console'           # Decoder currently supports console and wandb
+    data_path: str = './models'             # The path where files will be saved locally
+    init_config: Dict[str, Any] = None
+    wandb_entity: str = ''                  # Only needs to be set if tracker_type is wandb
+    wandb_project: str = ''
+    verbose: bool = False                   # Whether to print console logging for non-console trackers
+
+# diffusion prior pydantic classes
+
+class AdapterConfig(BaseModel):
+    make: str = "openai"
+    model: str = "ViT-L/14"
+    base_model_kwargs: Dict[str, Any] = None
+
+    def create(self):
+        if self.make == "openai":
+            return OpenAIClipAdapter(self.model)
+        elif self.make == "x-clip":
+            return XClipAdapter(XCLIP(**self.base_model_kwargs))
+        elif self.make == "coca":
+            return CoCaAdapter(CoCa(**self.base_model_kwargs))
+        else:
+            raise AttributeError("No adapter with that name is available.")
+
+class DiffusionPriorNetworkConfig(BaseModel):
+    dim: int
+    depth: int
+    num_timesteps: int = None
+    num_time_embeds: int = 1
+    num_image_embeds: int = 1
+    num_text_embeds: int = 1
+    dim_head: int = 64
+    heads: int = 8
+    ff_mult: int = 4
+    norm_out: bool = True
+    attn_dropout: float = 0.
+    ff_dropout: float = 0.
+    final_proj: bool = True
+    normformer: bool = False
+    rotary_emb: bool = True
+
+    def create(self):
+        kwargs = self.dict()
+        return DiffusionPriorNetwork(**kwargs)
+
+class DiffusionPriorConfig(BaseModel):
+    clip: AdapterConfig
+    net: DiffusionPriorNetworkConfig
+    image_embed_dim: int
+    image_size: int
+    image_channels: int = 3
+    timesteps: int = 1000
+    cond_drop_prob: float = 0.
+    loss_type: str = 'l2'
+    predict_x_start: bool = True
+    beta_schedule: str = 'cosine'
+    condition_on_text_encodings: bool = True
+
+    class Config:
+        extra = "allow"
+
+    def create(self):
+        kwargs = self.dict()
+        clip = AdapterConfig(**kwargs.pop('clip')).create()
+        diffusion_prior_network = DiffusionPriorNetworkConfig(**kwargs.pop('net')).create()
+        return DiffusionPrior(net = diffusion_prior_network, clip=clip, **kwargs)
+
+class DiffusionPriorTrainConfig(BaseModel):
+    epochs: int = 1
+    lr: float = 1.1e-4
+    wd: float = 6.02e-2
+    max_grad_norm: float = 0.5
+    use_ema: bool = True
+    ema_beta: float = 0.99
+    amp: bool = False
+    save_every: int = 10000 # what steps to save on
+
+class DiffusionPriorDataConfig(BaseModel):
+    image_url: str     # path to embeddings folder
+    meta_url: str      # path to metadata (captions) for images
+    splits: TrainSplitConfig
+    batch_size: int = 64
+
+class DiffusionPriorLoadConfig(BaseModel):
+    source: str = None
+    resume: bool = False
+
+class TrainDiffusionPriorConfig(BaseModel):
+    prior: DiffusionPriorConfig
+    data: DiffusionPriorDataConfig
+    train: DiffusionPriorTrainConfig
+    load: DiffusionPriorLoadConfig
+    tracker: TrackerConfig
+
+    @classmethod
+    def from_json_path(cls, json_path):
+        with open(json_path) as f:
+            config = json.load(f)
+        return cls(**config)
+
+# decoder pydantic classes
+
 class UnetConfig(BaseModel):
     dim: int
-    dim_mults: List[int]
+    dim_mults: ListOrTuple(int)
     image_embed_dim: int = None
     cond_dim: int = None
     channels: int = 3
@@ -21,13 +157,22 @@ class UnetConfig(BaseModel):
         extra = "allow"
 
 class DecoderConfig(BaseModel):
+    unets: ListOrTuple(UnetConfig)
     image_size: int = None
-    image_sizes: Union[List[int], Tuple[int]] = None
+    image_sizes: ListOrTuple(int) = None
     channels: int = 3
     timesteps: int = 1000
     loss_type: str = 'l2'
     beta_schedule: str = 'cosine'
     learned_variance: bool = True
+    image_cond_drop_prob: float = 0.1
+    text_cond_drop_prob: float = 0.5
+
+    def create(self):
+        decoder_kwargs = self.dict()
+        unet_configs = decoder_kwargs.pop('unets')
+        unets = [Unet(**config) for config in unet_configs]
+        return Decoder(unets, **decoder_kwargs)
 
     @validator('image_sizes')
     def check_image_sizes(cls, image_sizes, values):
@@ -38,17 +183,6 @@ class DecoderConfig(BaseModel):
     class Config:
         extra = "allow"
 
-class TrainSplitConfig(BaseModel):
-    train: float = 0.75
-    val: float = 0.15
-    test: float = 0.1
-
-    @root_validator
-    def validate_all(cls, fields):
-        if sum([*fields.values()]) != 1.:
-            raise ValueError(f'{fields.keys()} must sum to 1.0')
-        return fields
-
 class DecoderDataConfig(BaseModel):
     webdataset_base_url: str     # path to a webdataset with jpg images
     embeddings_url: str          # path to .npy files with embeddings
@@ -63,54 +197,6 @@ class DecoderDataConfig(BaseModel):
     resample_train: bool = False
     preprocessing: Dict[str, Any] = {'ToTensor': True}
 
-class DecoderTrainConfig(BaseModel):
-    epochs: int = 20
-    lr: float = 1e-4
-    wd: float = 0.01
-    max_grad_norm: float = 0.5
-    save_every_n_samples: int = 100000
-    n_sample_images: int = 6                # The number of example images to produce when sampling the train and test dataset
-    device: str = 'cuda:0'
-    epoch_samples: int = None               # Limits the number of samples per epoch. None means no limit. Required if resample_train is true as otherwise the number of samples per epoch is infinite.
-    validation_samples: int = None          # Same as above but for validation.
-    use_ema: bool = True
-    ema_beta: float = 0.99
-    amp: bool = False
-    save_all: bool = False                  # Whether to preserve all checkpoints
-    save_latest: bool = True                # Whether to always save the latest checkpoint
-    save_best: bool = True                  # Whether to save the best checkpoint
-    unet_training_mask: List[bool] = None   # If None, use all unets
-
-class DecoderEvaluateConfig(BaseModel):
-    n_evaluation_samples: int = 1000
-    FID: Dict[str, Any] = None
-    IS: Dict[str, Any] = None
-    KID: Dict[str, Any] = None
-    LPIPS: Dict[str, Any] = None
-
-class TrackerConfig(BaseModel):
-    tracker_type: str = 'console'           # Decoder currently supports console and wandb
-    data_path: str = './models'             # The path where files will be saved locally
-    init_config: Dict[str, Any] = None
-    wandb_entity: str = ''                  # Only needs to be set if tracker_type is wandb
-    wandb_project: str = ''
-    verbose: bool = False                   # Whether to print console logging for non-console trackers
-
-class DecoderLoadConfig(BaseModel):
-    source: str = None                      # Supports file and wandb
-    run_path: str = ''                      # Used only if source is wandb
-    file_path: str = ''                     # The local filepath if source is file. If source is wandb, the relative path to the model file in wandb.
-    resume: bool = False                    # If using wandb, whether to resume the run
-
-class TrainDecoderConfig(BaseModel):
-    unets: List[UnetConfig]
-    decoder: DecoderConfig
-    data: DecoderDataConfig
-    train: DecoderTrainConfig
-    evaluate: DecoderEvaluateConfig
-    tracker: TrackerConfig
-    load: DecoderLoadConfig
-
     @property
     def img_preproc(self):
         def _get_transformation(transformation_name, **kwargs):
@@ -120,9 +206,54 @@ class TrainDecoderConfig(BaseModel):
                 return T.RandomHorizontalFlip()
             elif transformation_name == "ToTensor":
                 return T.ToTensor()
-        
+
         transforms = []
-        for transform_name, transform_kwargs_or_bool in self.data.preprocessing.items():
+        for transform_name, transform_kwargs_or_bool in self.preprocessing.items():
             transform_kwargs = {} if not isinstance(transform_kwargs_or_bool, dict) else transform_kwargs_or_bool
             transforms.append(_get_transformation(transform_name, **transform_kwargs))
         return T.Compose(transforms)
+
+class DecoderTrainConfig(BaseModel):
+    epochs: int = 20
+    lr: float = 1e-4
+    wd: float = 0.01
+    max_grad_norm: float = 0.5
+    save_every_n_samples: int = 100000
+    n_sample_images: int = 6                       # The number of example images to produce when sampling the train and test dataset
+    device: str = 'cuda:0'
+    epoch_samples: int = None                      # Limits the number of samples per epoch. None means no limit. Required if resample_train is true as otherwise the number of samples per epoch is infinite.
+    validation_samples: int = None                 # Same as above but for validation.
+    use_ema: bool = True
+    ema_beta: float = 0.99
+    amp: bool = False
+    save_all: bool = False                         # Whether to preserve all checkpoints
+    save_latest: bool = True                       # Whether to always save the latest checkpoint
+    save_best: bool = True                         # Whether to save the best checkpoint
+    unet_training_mask: ListOrTuple(bool) = None   # If None, use all unets
+
+class DecoderEvaluateConfig(BaseModel):
+    n_evaluation_samples: int = 1000
+    FID: Dict[str, Any] = None
+    IS: Dict[str, Any] = None
+    KID: Dict[str, Any] = None
+    LPIPS: Dict[str, Any] = None
+
+class DecoderLoadConfig(BaseModel):
+    source: str = None                      # Supports file and wandb
+    run_path: str = ''                      # Used only if source is wandb
+    file_path: str = ''                     # The local filepath if source is file. If source is wandb, the relative path to the model file in wandb.
+    resume: bool = False                    # If using wandb, whether to resume the run
+
+class TrainDecoderConfig(BaseModel):
+    decoder: DecoderConfig
+    data: DecoderDataConfig
+    train: DecoderTrainConfig
+    evaluate: DecoderEvaluateConfig
+    tracker: TrackerConfig
+    load: DecoderLoadConfig
+
+    @classmethod
+    def from_json_path(cls, json_path):
+        with open(json_path) as f:
+            config = json.load(f)
+        return cls(**config)

dalle2_pytorch/trainer.py

@@ -1,5 +1,6 @@
 import time
 import copy
+from pathlib import Path
 from math import ceil
 from functools import partial, wraps
 from collections.abc import Iterable
@@ -55,6 +56,10 @@ def num_to_groups(num, divisor):
         arr.append(remainder)
     return arr
 
+def get_pkg_version():
+    from pkg_resources import get_distribution
+    return get_distribution('dalle2_pytorch').version
+
 # decorators
 
 def cast_torch_tensor(fn):
@@ -128,12 +133,6 @@ def split_args_and_kwargs(*args, split_size = None, **kwargs):
         chunk_size_frac = chunk_size / batch_size
         yield chunk_size_frac, (chunked_args, chunked_kwargs)
 
-# print helpers
-
-def print_ribbon(s, symbol = '=', repeat = 40):
-    flank = symbol * repeat
-    return f'{flank} {s} {flank}'
-
 # saving and loading functions
 
 # for diffusion prior
@@ -191,7 +190,7 @@ class EMA(nn.Module):
         self.update_after_step = update_after_step  // update_every # only start EMA after this step number, starting at 0
 
         self.register_buffer('initted', torch.Tensor([False]))
-        self.register_buffer('step', torch.tensor([0.]))
+        self.register_buffer('step', torch.tensor([0]))
 
     def restore_ema_model_device(self):
         device = self.initted.device
@@ -255,6 +254,7 @@ class DiffusionPriorTrainer(nn.Module):
         eps = 1e-6,
         max_grad_norm = None,
         amp = False,
+        group_wd_params = True,
         **kwargs
     ):
         super().__init__()
@@ -280,6 +280,7 @@ class DiffusionPriorTrainer(nn.Module):
             lr = lr,
             wd = wd,
             eps = eps,
+            group_wd_params = group_wd_params,
             **kwargs
         )
 
@@ -287,7 +288,50 @@ class DiffusionPriorTrainer(nn.Module):
 
         self.max_grad_norm = max_grad_norm
 
-        self.register_buffer('step', torch.tensor([0.]))
+        self.register_buffer('step', torch.tensor([0]))
+
+    def save(self, path, overwrite = True, **kwargs):
+        path = Path(path)
+        assert not (path.exists() and not overwrite)
+        path.parent.mkdir(parents = True, exist_ok = True)
+
+        save_obj = dict(
+            scaler = self.scaler.state_dict(),
+            optimizer = self.optimizer.state_dict(),
+            model = self.diffusion_prior.state_dict(),
+            version = get_pkg_version(),
+            step = self.step.item(),
+            **kwargs
+        )
+
+        if self.use_ema:
+            save_obj = {**save_obj, 'ema': self.ema_diffusion_prior.state_dict()}
+
+        torch.save(save_obj, str(path))
+
+    def load(self, path, only_model = False, strict = True):
+        path = Path(path)
+        assert path.exists()
+
+        loaded_obj = torch.load(str(path))
+
+        if get_pkg_version() != loaded_obj['version']:
+            print(f'loading saved diffusion prior at version {loaded_obj["version"]} but current package version is at {get_pkg_version()}')
+
+        self.diffusion_prior.load_state_dict(loaded_obj['model'], strict = strict)
+        self.step.copy_(torch.ones_like(self.step) * loaded_obj['step'])
+
+        if only_model:
+            return loaded_obj
+
+        self.scaler.load_state_dict(loaded_obj['scaler'])
+        self.optimizer.load_state_dict(loaded_obj['optimizer'])
+
+        if self.use_ema:
+            assert 'ema' in loaded_obj
+            self.ema_diffusion_prior.load_state_dict(loaded_obj['ema'], strict = strict)
+
+        return loaded_obj
 
     def update(self):
         if exists(self.max_grad_norm):
@@ -368,6 +412,7 @@ class DecoderTrainer(nn.Module):
         eps = 1e-8,
         max_grad_norm = 0.5,
         amp = False,
+        group_wd_params = True,
         **kwargs
     ):
         super().__init__()
@@ -393,6 +438,7 @@ class DecoderTrainer(nn.Module):
                 lr = unet_lr,
                 wd = unet_wd,
                 eps = unet_eps,
+                group_wd_params = group_wd_params,
                 **kwargs
             )
 
@@ -410,6 +456,60 @@ class DecoderTrainer(nn.Module):
 
         self.register_buffer('step', torch.tensor([0.]))
 
+    def save(self, path, overwrite = True, **kwargs):
+        path = Path(path)
+        assert not (path.exists() and not overwrite)
+        path.parent.mkdir(parents = True, exist_ok = True)
+
+        save_obj = dict(
+            model = self.decoder.state_dict(),
+            version = get_pkg_version(),
+            step = self.step.item(),
+            **kwargs
+        )
+
+        for ind in range(0, self.num_unets):
+            scaler_key = f'scaler{ind}'
+            optimizer_key = f'scaler{ind}'
+            scaler = getattr(self, scaler_key)
+            optimizer = getattr(self, optimizer_key)
+            save_obj = {**save_obj, scaler_key: scaler.state_dict(), optimizer_key: optimizer.state_dict()}
+
+        if self.use_ema:
+            save_obj = {**save_obj, 'ema': self.ema_unets.state_dict()}
+
+        torch.save(save_obj, str(path))
+
+    def load(self, path, only_model = False, strict = True):
+        path = Path(path)
+        assert path.exists()
+
+        loaded_obj = torch.load(str(path))
+
+        if get_pkg_version() != loaded_obj['version']:
+            print(f'loading saved decoder at version {loaded_obj["version"]}, but current package version is {get_pkg_version()}')
+
+        self.decoder.load_state_dict(loaded_obj['model'], strict = strict)
+        self.step.copy_(torch.ones_like(self.step) * loaded_obj['step'])
+
+        if only_model:
+            return loaded_obj
+
+        for ind in range(0, self.num_unets):
+            scaler_key = f'scaler{ind}'
+            optimizer_key = f'scaler{ind}'
+            scaler = getattr(self, scaler_key)
+            optimizer = getattr(self, optimizer_key)
+
+            scaler.load_state_dict(loaded_obj[scaler_key])
+            optimizer.load_state_dict(loaded_obj[optimizer_key])
+
+        if self.use_ema:
+            assert 'ema' in loaded_obj
+            self.ema_unets.load_state_dict(loaded_obj['ema'], strict = strict)
+
+        return loaded_obj
+
     @property
     def unets(self):
         return nn.ModuleList([ema.ema_model for ema in self.ema_unets])

dalle2_pytorch/utils.py

@@ -1,5 +1,7 @@
 import time
 
+# time helpers
+
 class Timer:
     def __init__(self):
         self.reset()
@@ -9,3 +11,9 @@ class Timer:
 
     def elapsed(self):
         return time.time() - self.last_time
+
+# print helpers
+
+def print_ribbon(s, symbol = '=', repeat = 40):
+    flank = symbol * repeat
+    return f'{flank} {s} {flank}'

setup.py

@@ -10,7 +10,7 @@ setup(
       'dream = dalle2_pytorch.cli:dream'
     ],
   },
-  version = '0.4.1',
+  version = '0.5.3',
   license='MIT',
   description = 'DALL-E 2',
   author = 'Phil Wang',

train_decoder.py

@@ -1,11 +1,10 @@
 from dalle2_pytorch import Unet, Decoder
-from dalle2_pytorch.trainer import DecoderTrainer, print_ribbon
+from dalle2_pytorch.trainer import DecoderTrainer
 from dalle2_pytorch.dataloaders import create_image_embedding_dataloader
 from dalle2_pytorch.trackers import WandbTracker, ConsoleTracker
 from dalle2_pytorch.train_configs import TrainDecoderConfig
-from dalle2_pytorch.utils import Timer
+from dalle2_pytorch.utils import Timer, print_ribbon
 
-import json
 import torchvision
 import torch
 from torchmetrics.image.fid import FrechetInceptionDistance
@@ -86,20 +85,6 @@ def create_dataloaders(
         "test_sampling": test_sampling_dataloader
     }
 
-
-def create_decoder(device, decoder_config, unets_config):
-    """Creates a sample decoder"""
-
-    unets = [Unet(**config.dict()) for config in unets_config]
-
-    decoder = Decoder(
-        unet=unets,
-        **decoder_config.dict()
-    )
-
-    decoder.to(device=device)
-    return decoder
-
 def get_dataset_keys(dataloader):
     """
     It is sometimes neccesary to get the keys the dataloader is returning. Since the dataset is burried in the dataloader, we need to do a process to recover it.
@@ -421,7 +406,7 @@ def initialize_training(config):
 
     dataloaders = create_dataloaders (
         available_shards=all_shards,
-        img_preproc = config.img_preproc,
+        img_preproc = config.data.img_preproc,
         train_prop = config.data.splits.train,
         val_prop = config.data.splits.val,
         test_prop = config.data.splits.test,
@@ -429,7 +414,7 @@ def initialize_training(config):
         **config.data.dict()
     )
 
-    decoder = create_decoder(device, config.decoder, config.unets)
+    decoder = config.decoder.create().to(device = device)
     num_parameters = sum(p.numel() for p in decoder.parameters())
     print(print_ribbon("Loaded Config", repeat=40))
     print(f"Number of parameters: {num_parameters}")
@@ -449,9 +434,7 @@ def initialize_training(config):
 @click.option("--config_file", default="./train_decoder_config.json", help="Path to config file")
 def main(config_file):
     print("Recalling config from {}".format(config_file))
-    with open(config_file) as f:
-        config = json.load(f)
-    config = TrainDecoderConfig(**config)
+    config = TrainDecoderConfig.from_json_path(config_file)
     initialize_training(config)
 
 

train_diffusion_prior.py

@@ -9,10 +9,10 @@ from torch import nn
 
 from dalle2_pytorch.dataloaders import make_splits
 from dalle2_pytorch import DiffusionPrior, DiffusionPriorNetwork, OpenAIClipAdapter
-from dalle2_pytorch.trainer import DiffusionPriorTrainer, load_diffusion_model, save_diffusion_model, print_ribbon
+from dalle2_pytorch.trainer import DiffusionPriorTrainer, load_diffusion_model, save_diffusion_model
 
 from dalle2_pytorch.trackers import ConsoleTracker, WandbTracker
-from dalle2_pytorch.utils import Timer
+from dalle2_pytorch.utils import Timer, print_ribbon
 
 from embedding_reader import EmbeddingReader