Skip to content
Snippets Groups Projects
Commit 578a3833 authored by Tobias Seibel's avatar Tobias Seibel
Browse files

minor bug fixes

parent ce99d71b
No related branches found
No related tags found
No related merge requests found
......@@ -2,3 +2,5 @@
*/__pycache__
*/trained_ddpm
root
experiments
trainer/__pycache__
\ No newline at end of file
......@@ -62,7 +62,7 @@ class UnconditionalDataset(Dataset):
return len(self.df)
def __getitem__(self,idx):
path = self.df.iloc[idx].Filepaths
path = self.df.iloc[idx].Filepath
img = Image.open(path)
return self.transform(img),0
......
%% Cell type:code id: tags:
``` python
from trainer.train import *
from dataloader.load import *
from models.Framework import *
from models.all_unets import *
import torch
from torch import nn
```
%% Cell type:markdown id: tags:
# Prepare experiment
1. Choose Hyperparameter Settings
2. Run notebook on local maschine to generate experiment folder with the JSON files containing the settings
3. scp experiment folder to the HPC
4. Run Pipeline by adding following to batch file:
- Train Model: &emsp;&emsp;&emsp;&emsp;&emsp; `python main.py train "<absolute path of experiment folder in hpc>"`
- Sample Images: &emsp;&emsp;&emsp; `python main.py sample "<absolute path of experiment folder in hpc>"`
- Evaluate Model: &emsp;&emsp;&emsp; `python main.py evaluate "<absolute path of experiment folder in hpc>"`
%% Cell type:code id: tags:
``` python
import torch
####
# Settings
####
# Dataset path
datapath = "/work/lect0100/lhq_256"
# Experiment setup
run_name = 'batch_timesteps' # WANDB and experiment folder Name!
run_name = 'main_test0' # WANDB and experiment folder Name!
checkpoint = None #'model_epoch_8.pth' # Name of checkpoint pth file or None
experiment_path = '/work/lect0100/experiments_gonzalo/'+ run_name +'/'
experiment_path = "/work/lect0100/main_experiment/" + run_name +'/'
# Path to save generated experiment folder on local machine
local_path ="/Users/gonzalo/Desktop/" + run_name + '/settings'
local_path ="experiments/" + run_name + '/settings'
# Diffusion Model Settings
diffusion_steps = 200
image_size = 64
diffusion_steps = 500
image_size = 128
channels = 3
# Training
batchsize = 32
epochs = 30
store_iter = 1
eval_iter = 500
epochs = 20
store_iter = 5
eval_iter = 2
learning_rate = 0.0001
optimizername = "torch.optim.AdamW"
optimizer_params = None
verbose = True
# checkpoint = None #(If no checkpoint training, ie. random weights)
# Sampling
sample_size = 10
sample_size = 20
intermediate = False # True if you want to sample one image and all ist intermediate latents
# Evaluating
...
###
# Advanced Settings Dictionaries
###
meta_setting = dict(modelname = "UNet_Res",
dataset = "UnconditionalDataset",
framework = "DDPM",
trainloop_function = "ddpm_trainer",
sampling_function = 'ddpm_sampler',
evaluation_function = 'ddpm_evaluator',
batchsize = batchsize
)
dataset_setting = dict(fpath = datapath,
img_size = image_size,
frac =0.8,
skip_first_n = 0,
ext = ".png",
transform=True
)
model_setting = dict( n_channels=64,
fctr = [1,2,4,4,8],
time_dim=256,
attention = True,
)
"""
outdated
model_setting = dict( channels_in=channels,
channels_out =channels ,
activation='relu', # activation function. Options: {'relu', 'leakyrelu', 'selu', 'gelu', 'silu'/'swish'}
weight_init='he', # weight initialization. Options: {'he', 'torch'}
projection_features=64, # number of image features after first convolution layer
time_dim=batchsize, #dont chnage!!!
time_channels=diffusion_steps, # number of time channels #TODO same as diffusion steps?
num_stages=4, # number of stages in contracting/expansive path
stage_list=None, # specify number of features produced by stages
num_blocks=1, # number of ConvResBlock in each contracting/expansive path
num_groupnorm_groups=32, # number of groups used in Group Normalization inside a ConvResBlock
dropout=0.1, # drop-out to be applied inside a ConvResBlock
attention_list=None, # specify MHA pattern across stages
num_attention_heads=1,
)
"""
framework_setting = dict(
diffusion_steps = diffusion_steps, # dont change!!
out_shape = (channels,image_size,image_size), # dont change!!
noise_schedule = 'linear',
beta_1 = 1e-4,
beta_T = 0.02,
alpha_bar_lower_bound = 0.9,
var_schedule = 'same',
kl_loss = 'simplified',
recon_loss = 'nll',
)
training_setting = dict(
epochs = epochs,
store_iter = store_iter,
eval_iter = eval_iter,
optimizer_class=optimizername,
optimizer_params = optimizer_params,
#optimizer_params=dict(lr=learning_rate), # don't change!
learning_rate = learning_rate,
run_name=run_name,
checkpoint= checkpoint,
experiment_path = experiment_path,
verbose = verbose,
T_max = 5*10000, # cosine lr param
eta_min= 1e-5, # cosine lr param
T_max = 0.8*90000/32*150, # cosine lr param len(train_ds)/batchsize * total epochs to 0
eta_min= 1e-10, # cosine lr param
)
sampling_setting = dict(
checkpoint = checkpoint,
experiment_path = experiment_path,
batch_size = sample_size,
intermediate = intermediate
)
# TODO
evaluation_setting = dict(
checkpoint = checkpoint,
experiment_path = experiment_path,
)
```
%% Cell type:code id: tags:
``` python
import os
import json
f = local_path
if os.path.exists(f):
print("path already exists, pick a new name!")
print("break")
else:
print("create folder")
#os.mkdir(f)
os.makedirs(f, exist_ok=True)
print("folder created ")
with open(f+"/meta_setting.json","w+") as fp:
json.dump(meta_setting,fp)
with open(f+"/dataset_setting.json","w+") as fp:
json.dump(dataset_setting,fp)
with open(f+"/model_setting.json","w+") as fp:
json.dump(model_setting,fp)
with open(f+"/framework_setting.json","w+") as fp:
json.dump(framework_setting,fp)
with open(f+"/training_setting.json","w+") as fp:
json.dump(training_setting,fp)
with open(f+"/sampling_setting.json","w+") as fp:
json.dump(sampling_setting,fp)
with open(f+"/evaluation_setting.json","w+") as fp:
json.dump(evaluation_setting,fp)
print("stored json files in folder")
print(meta_setting)
print(dataset_setting)
print(model_setting)
print(framework_setting)
print(training_setting)
print(sampling_setting)
print(evaluation_setting)
```
%% Output
create folder
folder created
stored json files in folder
{'modelname': 'UNet_Unconditional_Diffusion_Bottleneck_Variant', 'dataset': 'UnconditionalDataset', 'framework': 'DDPM', 'trainloop_function': 'ddpm_trainer', 'sampling_function': 'ddpm_sampler', 'evaluation_function': 'ddpm_evaluator', 'batchsize': 32}
{'fpath': '/work/lect0100/lhq_256', 'img_size': 64, 'frac': 0.8, 'skip_first_n': 0, 'ext': '.png', 'transform': True}
{'channels_in': 3, 'channels_out': 3, 'activation': 'relu', 'weight_init': 'he', 'projection_features': 64, 'time_dim': 32, 'time_channels': 200, 'num_stages': 4, 'stage_list': None, 'num_blocks': 1, 'num_groupnorm_groups': 32, 'dropout': 0.1, 'attention_list': None, 'num_attention_heads': 1}
{'diffusion_steps': 200, 'out_shape': (3, 64, 64), 'noise_schedule': 'linear', 'beta_1': 0.0001, 'beta_T': 0.02, 'alpha_bar_lower_bound': 0.9, 'var_schedule': 'same', 'kl_loss': 'simplified', 'recon_loss': 'nll'}
{'epochs': 30, 'store_iter': 1, 'eval_iter': 500, 'optimizer_class': 'torch.optim.AdamW', 'optimizer_params': None, 'learning_rate': 0.0001, 'run_name': 'batch_timesteps', 'checkpoint': None, 'experiment_path': '/work/lect0100/experiments_gonzalo/batch_timesteps/', 'verbose': True}
{'checkpoint': None, 'experiment_path': '/work/lect0100/experiments_gonzalo/batch_timesteps/', 'batch_size': 10, 'intermediate': False}
{'checkpoint': None, 'experiment_path': '/work/lect0100/experiments_gonzalo/batch_timesteps/'}
%% Cell type:code id: tags:
``` python
```
{'modelname': 'UNet_Res', 'dataset': 'UnconditionalDataset', 'framework': 'DDPM', 'trainloop_function': 'ddpm_trainer', 'sampling_function': 'ddpm_sampler', 'evaluation_function': 'ddpm_evaluator', 'batchsize': 32}
{'fpath': '/work/lect0100/lhq_256', 'img_size': 128, 'frac': 0.8, 'skip_first_n': 0, 'ext': '.png', 'transform': True}
{'n_channels': 64, 'fctr': [1, 2, 4, 4, 8], 'time_dim': 256}
{'diffusion_steps': 500, 'out_shape': (3, 128, 128), 'noise_schedule': 'linear', 'beta_1': 0.0001, 'beta_T': 0.02, 'alpha_bar_lower_bound': 0.9, 'var_schedule': 'same', 'kl_loss': 'simplified', 'recon_loss': 'nll'}
{'epochs': 10, 'store_iter': 2, 'eval_iter': 2, 'optimizer_class': 'torch.optim.AdamW', 'optimizer_params': None, 'learning_rate': 0.0001, 'run_name': 'main_testing', 'checkpoint': None, 'experiment_path': '/work/lect0100/tobi/main_test/main_testing/', 'verbose': True, 'T_max': 9000000, 'eta_min': 1e-10}
{'checkpoint': None, 'experiment_path': '/work/lect0100/tobi/main_test/main_testing/', 'batch_size': 10, 'intermediate': False}
{'checkpoint': None, 'experiment_path': '/work/lect0100/tobi/main_test/main_testing/'}
%% Cell type:code id: tags:
``` python
```
......
%% Cell type:code id: tags:
```
``` python
from trainer.train import trainer
from dataloader.load import DiffusionLoader
from models.model import VanillaDiffusion
```
%% Cell type:code id: tags:
```
``` python
import matplotlib.pyplot as plt
import numpy as np
eta_min = 0.00001
eta_max = 0.001
T_max = 0.8*90000/32*20 # len(trainset)= 0.8*90.000 T_max = len(trainset)/batch_size*epochs <- in the last epochsm eta min will be reached.
x = np.arange(T_max)
y = eta_min+ 0.5*(eta_max-eta_min)*(1+np.cos(np.pi*x/T_max))
plt.plot(x,y)
print(y[6000]) # one epoch has about 2.4k steps -> in the third epoch this will be the loss
```
%% Output
0.0009572050015330874
%% Cell type:code id: tags:
``` python
```
......
......@@ -174,7 +174,7 @@ def ddpm_trainer(model,
ema = ModelEmaV2(model, decay=decay, device = model.device)
# Using W&B
with wandb.init(project='test-project', name=run_name, entity='gonzalomartingarcia0', id=run_name, resume=True) as run:
with wandb.init(project='Unconditional Landscapes', name=run_name, entity='deep-lab-', id=run_name, resume=True) as run:
# Log some info
run.config.learning_rate = learning_rate
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment