[
  [
    "sup",
    {
      "src": "en",
      "tgt": "si",
      "train": {
        "parameters": [
          "--ddp-backend no_c10d",
          "--arch transformer",
          "--share-all-embeddings",
          "--encoder-layers 5",
          "--decoder-layers 5",
          "--encoder-embed-dim 512",
          "--decoder-embed-dim 512",
          "--encoder-ffn-embed-dim 2048",
          "--decoder-ffn-embed-dim 2048",
          "--encoder-attention-heads 8",
          "--decoder-attention-heads 8",
          "--encoder-normalize-before",
          "--decoder-normalize-before",
          "--attention-dropout 0.3",
          "--relu-dropout 0.3",
          "--optimizer adam",
          "--adam-betas \"(0.9, 0.98)\"",
          "--lr-scheduler inverse_sqrt",
          "--warmup-init-lr 1e-7",
          "--warmup-updates 4000",
          "--lr 0.003",
          "--min-lr 1e-9",
          "--clip-norm 0.0",
          "--dropout 0.3",
          "--criterion label_smoothed_cross_entropy",
          "--label-smoothing 0.2",
          "--max-tokens 4000",
          "--seed 1",
          "--log-format json",
          "--log-interval 100",
          "--max-epoch 100",
          "--save-interval 1",
          "--no-epoch-checkpoints"
        ],
        "gpu": 4
      },
      "translate": {
        "lenpen": 1.5,
        "max_token": 30000,
        "mono": "data/mono/sien/mono.bpe.en"
      }
    }
  ],
  [
    "bt_iter1",
    {
      "src": "si",
      "tgt": "en",
      "train": {
        "parameters": [
          "--fp16",
          "--ddp-backend no_c10d",
          "--arch transformer",
          "--share-all-embeddings",
          "--encoder-layers 6",
          "--decoder-layers 6",
          "--encoder-embed-dim 1024",
          "--decoder-embed-dim 1024",
          "--encoder-ffn-embed-dim 4096",
          "--decoder-ffn-embed-dim 4096",
          "--encoder-attention-heads 16",
          "--decoder-attention-heads 16",
          "--encoder-normalize-before",
          "--decoder-normalize-before",
          "--attention-dropout 0.1",
          "--relu-dropout 0.1",
          "--optimizer adam",
          "--adam-betas \"(0.9, 0.98)\"",
          "--lr-scheduler inverse_sqrt",
          "--warmup-init-lr 1e-7",
          "--warmup-updates 4000",
          "--lr 0.001",
          "--min-lr 1e-9",
          "--clip-norm 0.0",
          "--dropout 0.1",
          "--criterion label_smoothed_cross_entropy",
          "--label-smoothing 0.2",
          "--upsample-primary 1",
          "--max-tokens 4000",
          "--seed 1",
          "--log-format json",
          "--log-interval 100",
          "--max-update 120000",
          "--save-interval-updates 2000",
          "--keep-interval-updates 1",
          "--no-epoch-checkpoints"
        ],
        "gpu": 8
      },
      "translate": {
        "lenpen": 2.5,
        "max_token": 20000,
        "mono": "data/mono/sien/mono.bpe.si"
      }
    }
  ],
  [
    "bt_iter2",
    {
      "src": "en",
      "tgt": "si",
      "train": {
        "parameters": [
          "--fp16",
          "--ddp-backend no_c10d",
          "--arch transformer",
          "--share-all-embeddings",
          "--encoder-layers 6",
          "--decoder-layers 6",
          "--encoder-embed-dim 1024",
          "--decoder-embed-dim 1024",
          "--encoder-ffn-embed-dim 4096",
          "--decoder-ffn-embed-dim 4096",
          "--encoder-attention-heads 4",
          "--decoder-attention-heads 4",
          "--encoder-normalize-before",
          "--decoder-normalize-before",
          "--attention-dropout 0.2",
          "--relu-dropout 0.2",
          "--optimizer adam",
          "--adam-betas \"(0.9, 0.98)\"",
          "--lr-scheduler inverse_sqrt",
          "--warmup-init-lr 1e-7",
          "--warmup-updates 4000",
          "--lr 0.0007",
          "--min-lr 1e-9",
          "--clip-norm 0.0",
          "--dropout 0.2",
          "--criterion label_smoothed_cross_entropy",
          "--label-smoothing 0.2",
          "--upsample-primary 7",
          "--max-tokens 4000",
          "--seed 1",
          "--log-format json",
          "--log-interval 100",
          "--max-update 90000",
          "--save-interval-updates 2000",
          "--keep-interval-updates 1",
          "--no-epoch-checkpoints"
        ],
        "gpu": 8
      },
      "translate": {
        "lenpen": 0.8,
        "max_token": 20000,
        "mono": "data/mono/sien/mono.bpe.en"
      }
    }
  ]
]