def Create_Input_Files():
# Create input files (along with word map)
create_input_files(
dataset='coco',
karpathy_json_path=
'/Users/tangruixiang/Desktop/Fairness_Dataset/code/COCO/Karpathy_split/coco/dataset.json',
image_folder='/Volumes/Yuening\ Passport/rxtang/coco/images',
captions_per_image=5,
min_word_freq=5,
output_folder='processed_data',
max_len=50)
def run_model(input_summary_file,
output_summary_file,
out_crop_path,
result_output,
gl_epochs,
crop_type=None,
file_name=None,
run_path_absolute=r'C:\DSSAT47',
glue_flag=1,
simulation_model='B'):
utils.create_input_files(input_summary_file, output_summary_file)
utils.create_xfile(os.path.join(output_summary_file, 'xfile.json'),
out_crop_path, crop_type, file_name)
for fn in os.listdir(out_crop_path):
if os.path.splitext(fn)[-1] in list(SUFFIXES.values()):
dssat = DSSAT(os.path.join(out_crop_path, fn), run_path_absolute)
dssat(result_output, gl_epochs, glue_flag, simulation_model)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(
dataset='flickr8k',
karpathy_json_path='/scratch/nikolai/dataset_flickr8k.json',
image_folder='/scratch/nikolai/Flickr8K',
captions_per_image=5,
min_word_freq=5,
output_folder='/scratch/nikolai/out_data',
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files aling with word map
create_input_files(dataset = 'coco', karpathy_json_path = './caption_datasets/dataset_coco.json', \
image_folder = './images', captions_per_image = 5, min_word_freq = 5, \
output_folder = './data', max_len = 50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(dataset='coco',
karpathy_json_path='../splits/dataset_coco.json',
image_folder='/datasets/COCO-2015',
captions_per_image=5,
min_word_freq=5,
output_folder='/datasets/home/50/650/agokhale/285project/a-PyTorch-Tutorial-to-Image-Captioning/data_generated',
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(dataset='coco',
karpathy_json_path='caption_data/dataset_coco.json',
image_folder='/caption data/',
captions_per_image=5,
min_word_freq=5,
output_folder='caption_data/',
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(dataset='flickr8kzh',
karpathy_json_path='../data/flickr8kzh.json',
image_folder='../data/flickr8k_images/',
captions_per_image=5,
min_word_freq=5,
output_folder='../prepared_data/',
max_len=50,
char_based=False)
from utils import create_input_files, train_word2vec_model
if __name__ == '__main__':
create_input_files(
csv_folder='./data',
output_folder='./outdata',
# sentence_limit=15,
# word_limit=20,
# min_word_count=5)
sentence_limit=40,
word_limit=200,
min_word_count=3,
label_columns=[
1, 2, 3, 4, 5
] # # 'news', 'is_relevant', 'Armed Assault', 'Bombing/Explosion', 'Kidnapping', 'Other'
)
train_word2vec_model(data_folder='./outdata', algorithm='skipgram')
from utils import txt_2_json, create_input_files
from config import *
import os.path
if __name__ == '__main__':
# convert custom txt caption to karpathy json format, only once
if not os.path.isfile(caption_json_path):
txt_2_json()
# Create input files (along with word map)
create_input_files()
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(dataset='coco',
karpathy_json_path='./crawlers/output/JSON.json',
image_folder='',
captions_per_image=5,
min_word_freq=3,
output_folder='./output',
max_len=50)
help="caption datasets type,i.e.,'coco', 'flickr8k', 'flickr30k' ")
parser.add_argument("--karpathy_json_path",
default='data/dataset_coco.json',
type=str,
required=True,
help="annotation json file' ")
parser.add_argument(
"--image_folder",
default='data/',
type=str,
required=True,
help="the directory containing the train2014 and val2014 image folders"
)
parser.add_argument("--output_folder",
default='data/',
type=str,
required=True,
help="path for saving processed data")
parser.add_argument("--captions_per_image", default=5, type=int)
parser.add_argument("--min_word_freq", default=5, type=int)
parser.add_argument("--max_len", default=50, type=int)
args = parser.parse_args()
create_input_files(dataset=args.dataset,
karpathy_json_path=args.karpathy_json_path,
image_folder=args.image_folder,
captions_per_image=args.captions_per_image,
min_word_freq=args.min_word_freq,
output_folder=args.output_folder,
max_len=args.max_len)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(dataset='coco',
karpathy_json_path='scratch/project/dataset_coco.json',
image_folder='scratch/project/',
captions_per_image=5,
min_word_freq=5,
output_folder='scratch/project/preprocessed_data/',
max_len=50)
from utils import create_input_files, train_word2vec_model
if __name__ == '__main__':
create_input_files(csv_folder='./yahoo_answers_csv',
output_folder='./data',
sentence_limit=15,
word_limit=20,
min_word_count=5)
train_word2vec_model(data_folder='./data',
algorithm='skipgram')
from utils import create_input_files
import _config
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(dataset=_config.dataset,
karpathy_json_path=_config.karpathy_json_path,
image_folder=_config.image_foldr_path,
captions_per_image=_config.captions_per_image,
min_word_freq=_config.min_word_freq,
output_folder=_config.output_folder,
max_len=_config.max_sentence_length)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
address = ''
create_input_files(dataset='coco',
karpathy_json_path=address +
'karpathy/dataset_coco.json',
image_folder=address,
captions_per_image=5,
min_word_freq=5,
output_folder=address + 'out',
max_len=50)
from utils import create_input_files
from imageio import imread
if __name__ == '__main__':
# img = imread('./datasets/Flicker8k_Dataset/2513260012_03d33305cf.jpg')
create_input_files('flickr8k',
'./datasets/dataset_flickr8k.json',
'./datasets/Flicker8k_Dataset',
captions_per_image=5,
min_word_freq=5,
output_folder='./datasets/caption_data',
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
trainCaptionPath = "../datasets/coco2014/trainval_coco2014_captions/captions_train2014.json"
valCaptionPath = "../datasets/coco2014/trainval_coco2014_captions/captions_val2014.json"
trainImagePath = "../datasets/coco2014/"
valImagePath = "../datasets/coco2014/2014/"
captionPath = "../scratch/dataset_coco.json"
create_input_files(dataset='coco',
karpathy_json_path=captionPath,
image_folder=trainImagePath,
captions_per_image=5,
min_word_freq=5,
output_folder='../scratch/caption data/',
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(
dataset='coco',
karpathy_json_path=
'/home/work/jiangshuai/image_caption/data/dataset_coco.json',
image_folder='/home/work/jiangshuai/image_caption/data/',
captions_per_image=5,
min_word_freq=5,
output_folder='./data/',
max_len=50)
from utils import create_input_files
max_caption_length = 50
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(
dataset='flickr8k',
karpathy_json_path="data/Flicker8k_Dataset/karpathy_flickr8k.json",
image_folder='data/Flicker8k_Dataset/Flicker8k_images/',
captions_per_image=5,
min_word_freq=5,
output_folder='data/',
max_len=max_caption_length,
bert_model_name='bert-base-cased')
from utils import create_input_files
import os
envget = os.environ.get
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(
dataset='coco',
karpathy_json_path=os.path.join(envget('HOME'),
'data/captiondata/dataset_coco.json'),
image_folder=os.path.join(envget('HOME'), 'data/captiondata/'),
captions_per_image=5,
min_word_freq=5,
output_folder=os.path.join(envget('HOME'), 'data/captiondata/'),
max_len=50)
#train = rd.read_input_file(c.TRAIN_PATH, c.EMOTION_HEADER, c.ATTRIBUTES, c.CLEAN)
#test = rd.read_input_file(c.TEST_PATH, c.EMOTION_HEADER, c.ATTRIBUTES, c.CLEAN)
train = pd.read_csv("./data/nlp_train.csv")
test = pd.read_csv("./data/nlp_test.csv")
train = train.dropna()
test = test.dropna()
train_temp = train.loc[:, ['anger', 'body']]
test_temp = test.loc[:, ['anger', 'body']]
train_temp.to_csv('./data/train.csv', index=False, header=False)
test_temp.to_csv('./data/test.csv', index=False, header=False)
create_input_files(csv_folder='./data',
output_folder='./outdata',
# sentence_limit=15,
# word_limit=20,
# min_word_count=5)
sentence_limit=30,
word_limit=100,
min_word_count=10)
train_word2vec_model(data_folder='./outdata', algorithm='skipgram')
file1 = open("label.txt", "a")
file2 = open("result.txt", "a")
file2.close()
for i in c.LABELS:
print(i)
file1.write(i)
file1.write('\n')
os.system('python3 train.py')
os.system('python3 eval.py')
from utils import create_input_files
if __name__ == '__main__':
create_input_files(dataset = 'flickr8k',
karpathy_json_path = '../../Datasets/train_valid_test_splits/dataset_flickr8k.json',
image_folder = '../../Datasets/Flickr8K/Flicker8k_Dataset/',
captions_per_image = 5,
min_word_freq = 5,
ouput_folder = './data') # This will be in /netscratch/deshmukh
from utils import create_input_files
import os
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(
dataset='flickr8k',
karpathy_json_path=os.path.join(
"dataset_flickr8k.json"
), #'caption_datasets/dataset_flicker8k.json',
image_folder=os.path.join(
"Flicker8k_Dataset"), #'Flicker8k_Dataset/Flicker8k_Dataset/',
captions_per_image=5,
min_word_freq=5,
output_folder=os.path.join("Flicker8k_Dataset"),
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
json_path = '/home/jithin/datasets/imageCaptioning/captions/dataset_flickr8k.json'
img_folder = '/home/jithin/datasets/imageCaptioning/flicker8k/Flicker8k_Dataset'
output_folder = './pre_processed'
create_input_files(dataset='flickr8k',
karpathy_json_path=json_path,
image_folder=img_folder,
captions_per_image=5,
min_word_freq=5,
output_folder=output_folder,
max_len=50)
from utils import create_input_files
from macro import IMAGE_FOLDER, OUTPUT_FOLDER
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(
dataset='flickr30k', # 'coco',
karpathy_json_path=
'karpathy_split_json_path/dataset_flickr30k.json', # where data split is divided
image_folder=IMAGE_FOLDER, # '/media/ssd/caption data/',
captions_per_image=5,
min_word_freq=5,
output_folder=OUTPUT_FOLDER, # ''/media/ssd/caption data/',
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(
dataset='flickr30k',
karpathy_json_path='../data/dataset_flickr30k.json',
image_folder='../data/flickr30k_images/flickr30k_images/',
captions_per_image=5,
min_word_freq=5,
output_folder='../data',
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(dataset='atlas',
karpathy_json_path='../../dataset/atlas_dataset.json',
image_folder='../../dataset/',
captions_per_image=1,
min_word_freq=1,
output_folder='output',
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(dataset='flickr8k',
karpathy_json_path='dataset_flickr8k.json',
image_folder='Flickr8k_Dataset/gaussian_0.13',
captions_per_image=5,
min_word_freq=5,
output_folder='dataset_gaussian_0.13',
max_len=50)
from utils import create_input_files
import argparse
import os
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='Create input files')
parser.add_argument('--dataset', '-d', default='coco', help='dataset')
parser.add_argument('--img_folder',
'-i',
default='image_folder',
help='path to image')
parser.add_argument('--caption_folder',
'-cf',
default='caption_datasets',
help='path to captions')
args = parser.parse_args()
# Create input files (along with word map)
create_input_files(dataset=args.dataset,
karpathy_json_path=os.path.join(
args.caption_folder,
'dataset_{:s}.json'.format(args.dataset)),
image_folder=args.img_folder,
captions_per_image=5,
min_word_freq=5,
output_folder='{:s}_folder'.format(args.dataset),
max_len=50)
from utils import create_input_files
if __name__ == '__main__':
# Create input files (along with word map)
create_input_files(
dataset='face2text',
karpathy_json_path='./caption data/dataset_face2text.json',
image_folder='./caption data/dataset-face2text/',
captions_per_image=2,
min_word_freq=2,
output_folder='./caption data/',
max_len=50)
