def main():
file = open(GOOGLE_CAPTIONS_FILE, 'r')
# Ensure workspace directory exists
make_directories(WORKSPACE)
index_file = open(os.path.join(WORKSPACE, "index_file.txt"), 'a')
count = 0
future_list = list()
with concurrent.futures.ThreadPoolExecutor(max_workers=15) as executor:
for line in file:
count += 1
future_list.append(
executor.submit(download_image,
line.strip().split("\t")[-1]))
if count >= NUM_SAMPLES:
break
for future in concurrent.futures.as_completed(future_list):
try:
url, file_name = future.result()
except Exception as exc:
print("Executor Service exception..!")
else:
if url is not None:
index_file.write(url + " " + file_name + "\n")
nargs="?",
type=str,
default="../configs/resnet50_lstm_with_threshold.yaml",
help="Configuration file to use",
)
args = parser.parse_args()
with open(args.config) as fp:
cfg = yaml.load(fp)
dataset_cfg = get_dataset_metadata_cfg()
model_workspace_dir = os.path.join(cfg["workspace"]["directory"],
cfg["dataset"]["name"],
cfg["model"]["arch"])
utils.make_directories(model_workspace_dir)
img_model = ResNet50(weights='imagenet')
dataset_preprocessor = PreProcessing(cfg, "resnet50", False, False)
dataset_preprocessor.run_one_time_encoding(img_model)
# Load train, validation sets from the pre-processor
training_generator, validation_generator, test_generator = dataset_preprocessor.get_keras_generators(
"resnet50")
MAX_LEN = 40
EMBEDDING_DIM = 300
IMAGE_ENC_DIM = 300
vocab_size = get_line_count(
os.path.join(cfg["workspace"]["directory"], cfg["dataset"]["name"],