add MT model

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 10.0,
+    "predict_bleu": 29.283,
+    "predict_gen_len": 31.4311,
+    "predict_loss": 3.1916561126708984,
+    "predict_runtime": 649.2349,
+    "predict_samples": 1835,
+    "predict_samples_per_second": 2.826,
+    "predict_steps_per_second": 0.283,
+    "train_loss": 0.5166526954256896,
+    "train_runtime": 10734.5111,
+    "train_samples": 30782,
+    "train_samples_per_second": 28.676,
+    "train_steps_per_second": 2.868
+}

config.json

@@ -0,0 +1,63 @@
+{
+  "_name_or_path": "/data/users/didelani/afribart/afrimbart",
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "gelu",
+  "adapters": {
+    "adapters": {},
+    "config_map": {},
+    "fusion_config_map": {},
+    "fusions": {}
+  },
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 250020,
+  "dropout": 0.1,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250004,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_length": 1024,
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_beams": 5,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.10.3",
+  "use_cache": true,
+  "vocab_size": 250054
+}

predict_results.json

@@ -0,0 +1,9 @@
+{
+    "predict_bleu": 29.283,
+    "predict_gen_len": 31.4311,
+    "predict_loss": 3.1916561126708984,
+    "predict_runtime": 649.2349,
+    "predict_samples": 1835,
+    "predict_samples_per_second": 2.826,
+    "predict_steps_per_second": 0.283
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e41050d313c02c661de97356324ba5d185010c1d3d6a968b9980c465a21909c
+size 2444723641

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>", "additional_special_tokens": ["ar_AR", "cs_CZ", "de_DE", "en_XX", "es_XX", "et_EE", "fi_FI", "fr_XX", "gu_IN", "hi_IN", "it_IT", "ja_XX", "kk_KZ", "ko_KR", "lt_LT", "lv_LV", "my_MM", "ne_NP", "nl_XX", "ro_RO", "ru_RU", "si_LK", "tr_TR", "vi_VN", "zh_CN", "af_ZA", "az_AZ", "bn_IN", "fa_IR", "he_IL", "hr_HR", "id_ID", "ka_GE", "km_KH", "mk_MK", "ml_IN", "mn_MN", "mr_IN", "pl_PL", "ps_AF", "pt_XX", "sv_SE", "sw_KE", "ta_IN", "te_IN", "th_TH", "tl_XX", "uk_UA", "ur_PK", "xh_ZA", "gl_ES", "sl_SI"]}

tokenizer_config.json

@@ -0,0 +1 @@
+{"src_lang": null, "tgt_lang": null, "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": "<s>", "model_max_length": 1024, "name_or_path": "/data/users/didelani/afribart/afrimbart", "special_tokens_map_file": "/home/suraj/projects/mbart-50/hf_models/mbart-50-large/special_tokens_map.json", "additional_special_tokens": ["ar_AR", "cs_CZ", "de_DE", "en_XX", "es_XX", "et_EE", "fi_FI", "fr_XX", "gu_IN", "hi_IN", "it_IT", "ja_XX", "kk_KZ", "ko_KR", "lt_LT", "lv_LV", "my_MM", "ne_NP", "nl_XX", "ro_RO", "ru_RU", "si_LK", "tr_TR", "vi_VN", "zh_CN", "af_ZA", "az_AZ", "bn_IN", "fa_IR", "he_IL", "hr_HR", "id_ID", "ka_GE", "km_KH", "mk_MK", "ml_IN", "mn_MN", "mr_IN", "pl_PL", "ps_AF", "pt_XX", "sv_SE", "sw_KE", "ta_IN", "te_IN", "th_TH", "tl_XX", "uk_UA", "ur_PK", "xh_ZA", "gl_ES", "sl_SI"], "sp_model_kwargs": {}, "tokenizer_class": "MBart50Tokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 0.5166526954256896,
+    "train_runtime": 10734.5111,
+    "train_samples": 30782,
+    "train_samples_per_second": 28.676,
+    "train_steps_per_second": 2.868
+}

trainer_state.json

@@ -0,0 +1,391 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "global_step": 30790,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.91880480675544e-05,
+      "loss": 2.5859,
+      "step": 500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.8376096135108804e-05,
+      "loss": 2.0293,
+      "step": 1000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.7564144202663205e-05,
+      "loss": 1.8726,
+      "step": 1500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.6752192270217606e-05,
+      "loss": 1.8068,
+      "step": 2000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.5940240337772006e-05,
+      "loss": 1.7605,
+      "step": 2500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.512828840532641e-05,
+      "loss": 1.9469,
+      "step": 3000
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.431633647288081e-05,
+      "loss": 1.2935,
+      "step": 3500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 4.350438454043521e-05,
+      "loss": 1.2342,
+      "step": 4000
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 4.269243260798961e-05,
+      "loss": 1.2234,
+      "step": 4500
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.188048067554401e-05,
+      "loss": 1.2641,
+      "step": 5000
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 4.106852874309841e-05,
+      "loss": 1.2556,
+      "step": 5500
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.025657681065281e-05,
+      "loss": 1.2404,
+      "step": 6000
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 3.944462487820721e-05,
+      "loss": 1.0196,
+      "step": 6500
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 3.8632672945761614e-05,
+      "loss": 0.8039,
+      "step": 7000
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 3.7820721013316015e-05,
+      "loss": 0.8155,
+      "step": 7500
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 3.7008769080870415e-05,
+      "loss": 0.8213,
+      "step": 8000
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 3.6196817148424816e-05,
+      "loss": 0.8316,
+      "step": 8500
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 3.538486521597922e-05,
+      "loss": 0.8499,
+      "step": 9000
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 3.457291328353362e-05,
+      "loss": 0.64,
+      "step": 9500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 3.376096135108802e-05,
+      "loss": 0.4628,
+      "step": 10000
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 3.294900941864242e-05,
+      "loss": 0.4861,
+      "step": 10500
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 3.213705748619682e-05,
+      "loss": 0.4887,
+      "step": 11000
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 3.132510555375122e-05,
+      "loss": 0.5098,
+      "step": 11500
+    },
+    {
+      "epoch": 3.9,
+      "learning_rate": 3.051315362130562e-05,
+      "loss": 0.4998,
+      "step": 12000
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 2.970120168886002e-05,
+      "loss": 0.4177,
+      "step": 12500
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 2.888924975641442e-05,
+      "loss": 0.257,
+      "step": 13000
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 2.8077297823968824e-05,
+      "loss": 0.264,
+      "step": 13500
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 2.7265345891523225e-05,
+      "loss": 0.278,
+      "step": 14000
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 2.6453393959077623e-05,
+      "loss": 0.2836,
+      "step": 14500
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 2.5641442026632023e-05,
+      "loss": 0.2851,
+      "step": 15000
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 2.4829490094186424e-05,
+      "loss": 0.2511,
+      "step": 15500
+    },
+    {
+      "epoch": 5.2,
+      "learning_rate": 2.401753816174083e-05,
+      "loss": 0.1468,
+      "step": 16000
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 2.3205586229295226e-05,
+      "loss": 0.1497,
+      "step": 16500
+    },
+    {
+      "epoch": 5.52,
+      "learning_rate": 2.2393634296849627e-05,
+      "loss": 0.1519,
+      "step": 17000
+    },
+    {
+      "epoch": 5.68,
+      "learning_rate": 2.158168236440403e-05,
+      "loss": 0.15,
+      "step": 17500
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 2.0769730431958428e-05,
+      "loss": 0.1559,
+      "step": 18000
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 1.995777849951283e-05,
+      "loss": 0.1542,
+      "step": 18500
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 1.914582656706723e-05,
+      "loss": 0.0818,
+      "step": 19000
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 1.833387463462163e-05,
+      "loss": 0.0873,
+      "step": 19500
+    },
+    {
+      "epoch": 6.5,
+      "learning_rate": 1.752192270217603e-05,
+      "loss": 0.0857,
+      "step": 20000
+    },
+    {
+      "epoch": 6.66,
+      "learning_rate": 1.6709970769730432e-05,
+      "loss": 0.0888,
+      "step": 20500
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 1.5898018837284833e-05,
+      "loss": 0.0853,
+      "step": 21000
+    },
+    {
+      "epoch": 6.98,
+      "learning_rate": 1.5086066904839236e-05,
+      "loss": 0.0862,
+      "step": 21500
+    },
+    {
+      "epoch": 7.15,
+      "learning_rate": 1.4274114972393635e-05,
+      "loss": 0.0554,
+      "step": 22000
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 1.3462163039948036e-05,
+      "loss": 0.0507,
+      "step": 22500
+    },
+    {
+      "epoch": 7.47,
+      "learning_rate": 1.2650211107502435e-05,
+      "loss": 0.0506,
+      "step": 23000
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 1.1838259175056837e-05,
+      "loss": 0.0507,
+      "step": 23500
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 1.1026307242611238e-05,
+      "loss": 0.0509,
+      "step": 24000
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 1.0214355310165639e-05,
+      "loss": 0.0487,
+      "step": 24500
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 9.40240337772004e-06,
+      "loss": 0.0351,
+      "step": 25000
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 8.59045144527444e-06,
+      "loss": 0.0304,
+      "step": 25500
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 7.778499512828841e-06,
+      "loss": 0.0306,
+      "step": 26000
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 6.966547580383241e-06,
+      "loss": 0.0297,
+      "step": 26500
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 6.154595647937642e-06,
+      "loss": 0.0292,
+      "step": 27000
+    },
+    {
+      "epoch": 8.93,
+      "learning_rate": 5.342643715492044e-06,
+      "loss": 0.0285,
+      "step": 27500
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 4.530691783046444e-06,
+      "loss": 0.0224,
+      "step": 28000
+    },
+    {
+      "epoch": 9.26,
+      "learning_rate": 3.718739850600845e-06,
+      "loss": 0.0188,
+      "step": 28500
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 2.9067879181552453e-06,
+      "loss": 0.0183,
+      "step": 29000
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 2.094835985709646e-06,
+      "loss": 0.018,
+      "step": 29500
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 1.2828840532640467e-06,
+      "loss": 0.0172,
+      "step": 30000
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 4.7093212081844755e-07,
+      "loss": 0.0171,
+      "step": 30500
+    },
+    {
+      "epoch": 10.0,
+      "step": 30790,
+      "total_flos": 8.428451462293094e+16,
+      "train_loss": 0.5166526954256896,
+      "train_runtime": 10734.5111,
+      "train_samples_per_second": 28.676,
+      "train_steps_per_second": 2.868
+    }
+  ],
+  "max_steps": 30790,
+  "num_train_epochs": 10,
+  "total_flos": 8.428451462293094e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a5785d08a7fae86b44a39783cf9d95e7702ee0287d60fa9cf9d86ff2fc401c1
+size 2799