{ "architecture": "show_attend_tell", "encoder": "resnet101_imagenet_frozen", "encoded_image_size": 14, "encoder_dim": 2048, "embed_dim": 256, "attention_dim": 256, "decoder_dim": 512, "dropout": 0.5, "vocab_size": 1605, "image_size": 224, "max_caption_len": 400, "min_word_freq": 2, "language": "km", "trained_on": "phonsobon/khmer_images_captioning_v2" }