prakashjayy

These are the list of papers concepts you should learn to become a PRO in GAN Research. We will see one as a step by step improvment in generation quality, stabilized training etc.

1. The first paper on GAN by IAN GoodFellow and team. There are many blogs on how a GAN works based on this paper. A simple search on google should give u many results. Asking an LLM is far more better.
2. GANs more susceptible to mode-collapse, vanishing gradients, so people moved away from simple BCE loss to Earth mover distance and some kind of regalurization to avoid these. Read WGAN and WGAN-GP papers to understand this again.
3. Next up is ProGAN. this is the first paper to my knowledge which basically made GANs possible to generate high resolution images. though progressive training is out of fashion, I still consider one should read about this [https://arxiv.org/pdf/1710.10196v3]. Their another key contribution is

prakashjayy

def get_tile_name_path(dst_dir, index, city=city, code=code):
    '''
    generating specific tile name
    '''
    dst_tile_name = "{}_{}_{}.tif".format(city, code, str(index).zfill(5))
    dst_tile_path = os.path.join(dst_dir, dst_tile_name)
    return dst_tile_name, dst_tile_path

def get_tile_transform(parent_transform, pixel_x,pixel_y):
    '''

prakashjayy

@patch_to(ds.FeatureCollection, cls_method=True)
def read_generic(cls, 
                 fp: str,
                 dst_crs: Union[str, rasterio.crs.CRS]=None,
):
    '''
    geojson reader and can also be used to ensure the geojson in mapped to a destined crs
    '''
    with open(fp, 'r', encoding='utf-8') as f:
        collection = json.load(f)

prakashjayy

# Fastscript usage under various instancees 
# https://github.com/fastai/fastscript/blob/master/00_core.ipynb
# https://fastcore.fast.ai/script.html

from fastcore.all import *

@call_parse 
def test_script(p: Param(help="any basic string", type=str)):
	print(len(p))

prakashjayy

import torch
import torch.nn as nn
import torch.optim as optim
from torch.optim import lr_scheduler
import numpy as np
import time
import os
import argparse

## Load the model 

prakashjayy

# This is the docker image available. I am using cpu version here. If needed there is gpu version available.
FROM bvlc/caffe:cpu

# Copy the file into docker
COPY requirements.txt requirements.txt

# Run the copied file
RUN pip install -r requirements.txt

# create a folder called model1 and copy all the files in the folder into that folder

prakashjayy

\begin{equation*}
\left[ \begin{array}{cccc}
i_{1} & i_{2} & i_{3} \\\end{array} \right] \times \left[ \begin{array}{cccc}
W_{i1j1} & W_{i1j2} & W_{i1j3} \\
W_{i2j1} & W_{i2j2} & W_{i2j3} \\
W_{i3j1} & W_{i3j2} & W_{i3j3} \\ \end{array} \right] = \left[ \begin{array}{cccc}
h_{1in} & h_{2in} & h_{3in} \\\end{array} \right]
\end{equation*}

prakashjayy

from keras import applications
from keras.preprocessing.image import ImageDataGenerator
from keras import optimizers
from keras.models import Sequential, Model 
from keras.layers import Dropout, Flatten, Dense, GlobalAveragePooling2D
from keras import backend as k 
from keras.callbacks import ModelCheckpoint, LearningRateScheduler, TensorBoard, EarlyStopping

img_width, img_height = 256, 256

prakashjayy

from keras import applications
from keras.preprocessing.image import ImageDataGenerator
from keras import optimizers
from keras.models import Sequential, Model 
from keras.layers import Dropout, Flatten, Dense, GlobalAveragePooling2D
from keras import backend as k 
from keras.callbacks import ModelCheckpoint, LearningRateScheduler, TensorBoard, EarlyStopping

img_width, img_height = 256, 256
train_data_dir = "data/train"