pandeydivesh15

n_input = images_train_norm.shape[1] # According to the paper, this must be equal to 32 * 32 = 1024
gbrbm_1 = GBRBM(n_input, 2000, learning_rate=0.001, use_tqdm=True, sigma=1)
# Fit image data in gbrbm_1.......
bbrbm_1 = BBRBM(2000, 1000, learning_rate=0.1, use_tqdm=True)
# Fit image data in bbrbm_1.......
bbrbm_2 = BBRBM(1000, 500, learning_rate=0.1, use_tqdm=True)
# Fit image data in bbrbm_2.......
bgrbm_1 = BGRBM(500, 50, learning_rate=0.001, use_tqdm=True, sigma=1)
# Fit image data in bgrbm_1.......

pandeydivesh15

{
  "transcript": "Here\nshe\nis\nnow\n.
  "words": [
    {
      "alignedWord": "here",
      "case": "success",
      "end": 49.339999999999996,
      "endOffset": 4,
      "phones": [
        {

pandeydivesh15

{"start": "13.03", "end": "13.37", "word": "trump"}
{"start": "13.37", "end": "13.75", "word": "pulls"}
{"start": "13.75", "end": "13.86", "word": "the"}
{"start": "13.86", "end": "14.30", "word": "trigger"}
{"start": "14.30", "end": "14.48", "word": "on"}
{"start": "14.51", "end": "14.77", "word": "two"}