src/visualize/text2motion.py

import os
import sys
sys.path.append('.')

import matplotlib.pyplot as plt
import torch
from src.utils.get_model_and_data import get_model_and_data
from src.parser.visualize import parser
from src.visualize.visualize import viz_clip_text, get_gpu_device
from src.utils.misc import load_model_wo_clip
import src.utils.fixseed  # noqa

plt.switch_backend('agg')


def main():
    # parse options
    parameters, folder, checkpointname, epoch = parser()
    gpu_device = get_gpu_device()
    parameters["device"] = f"cuda:{gpu_device}"
    model, datasets = get_model_and_data(parameters, split='vald')

    print("Restore weights..")
    checkpointpath = os.path.join(folder, checkpointname)
    state_dict = torch.load(checkpointpath, map_location=parameters["device"])
    load_model_wo_clip(model, state_dict)

    assert os.path.isfile(parameters['input_file'])
    with open(parameters['input_file'], 'r') as fr:
        texts = fr.readlines()
    texts = [s.replace('\n', '') for s in texts]
    grid = []
    for i in range(len(texts) // 4 + (len(texts) % 4 != 0)):
        grid.append(texts[i * 4:(i + 1) * 4])
    grid[-1] += [''] * (4 - len(grid[-1]))

    viz_clip_text(model, grid, epoch, parameters, folder=folder)


if __name__ == '__main__':
    main()