1st commit

README.md

@@ -0,0 +1,45 @@
+This is the [BSC-TeMU/roberta-large-bne](https://huggingface.co/BSC-TeMU/roberta-large-bne) model ([source](https://github.com/PlanTL-SANIDAD/lm-spanish)) trained on the [squad_es v2.0.0](https://huggingface.co/datasets/squad_es) dataset ([source](https://github.com/ccasimiro88/TranslateAlignRetrieve)).
+
+Current achievement: em=60.21, f1=68.61
+
+Results:
+
+```
+{
+    "epoch": 4.0,
+    "eval_HasAns_exact": 48.44804318488529,
+    "eval_HasAns_f1": 65.24520506718169,
+    "eval_HasAns_total": 5928,
+    "eval_NoAns_exact": 71.97301854974705,
+    "eval_NoAns_f1": 71.97301854974705,
+    "eval_NoAns_total": 5930,
+    "eval_best_exact": 60.22094788328555,
+    "eval_best_exact_thresh": 0.0,
+    "eval_best_f1": 68.6181122987237,
+    "eval_best_f1_thresh": 0.0,
+    "eval_exact": 60.2125147579693,
+    "eval_f1": 68.60967917340695,
+    "eval_samples": 12203,
+    "eval_total": 11858
+}
+```
+
+Training script:
+
+```
+python -m torch.distributed.launch --nproc_per_node=3 ./run_qa.py \
+    --model_name_or_path BSC-TeMU/roberta-large-bne \
+    --dataset_name squad_es \
+    --dataset_config_name v2.0.0 \
+    --do_train \
+    --do_eval \
+    --learning_rate 3e-5 \
+    --num_train_epochs 4 \
+    --max_seq_length 384 \
+    --doc_stride 128 \
+    --output_dir ./models/roberta-large-bne-finetuned-squad-es/ \
+    --per_device_eval_batch_size=24   \
+    --per_device_train_batch_size=12   \
+    --version_2_with_negative \
+    --ddp_find_unused_parameters=False \
+```

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 4.0,
+    "eval_HasAns_exact": 48.44804318488529,
+    "eval_HasAns_f1": 65.24520506718169,
+    "eval_HasAns_total": 5928,
+    "eval_NoAns_exact": 71.97301854974705,
+    "eval_NoAns_f1": 71.97301854974705,
+    "eval_NoAns_total": 5930,
+    "eval_best_exact": 60.22094788328555,
+    "eval_best_exact_thresh": 0.0,
+    "eval_best_f1": 68.6181122987237,
+    "eval_best_f1_thresh": 0.0,
+    "eval_exact": 60.2125147579693,
+    "eval_f1": 68.60967917340695,
+    "eval_samples": 12203,
+    "eval_total": 11858,
+    "train_loss": 0.7583477986664034,
+    "train_runtime": 15779.9103,
+    "train_samples": 132185,
+    "train_samples_per_second": 33.507,
+    "train_steps_per_second": 0.931
+}

config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "BSC-TeMU/roberta-large-bne",
+  "architectures": [
+    "RobertaForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.10.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50262
+}

eval_nbest_predictions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38d96f674136bc7d555542309b918e6325cefc40d7e3bb82b9df66c4689844e4
+size 56937102

eval_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 4.0,
+    "eval_HasAns_exact": 48.44804318488529,
+    "eval_HasAns_f1": 65.24520506718169,
+    "eval_HasAns_total": 5928,
+    "eval_NoAns_exact": 71.97301854974705,
+    "eval_NoAns_f1": 71.97301854974705,
+    "eval_NoAns_total": 5930,
+    "eval_best_exact": 60.22094788328555,
+    "eval_best_exact_thresh": 0.0,
+    "eval_best_f1": 68.6181122987237,
+    "eval_best_f1_thresh": 0.0,
+    "eval_exact": 60.2125147579693,
+    "eval_f1": 68.60967917340695,
+    "eval_samples": 12203,
+    "eval_total": 11858
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e51e102b24d1f25ef5bc157fe0558eea02e8bf2498bf49c66b96f1badd40053f
+size 1417399857

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": true, "errors": "replace", "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "max_len": 512, "special_tokens_map_file": null, "name_or_path": "BSC-TeMU/roberta-large-bne", "tokenizer_class": "RobertaTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "train_loss": 0.7583477986664034,
+    "train_runtime": 15779.9103,
+    "train_samples": 132185,
+    "train_samples_per_second": 33.507,
+    "train_steps_per_second": 0.931
+}

trainer_state.json

@@ -0,0 +1,199 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "global_step": 14688,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.14,
+      "learning_rate": 2.897875816993464e-05,
+      "loss": 2.0084,
+      "step": 500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 2.795751633986928e-05,
+      "loss": 1.5875,
+      "step": 1000
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.6936274509803924e-05,
+      "loss": 1.5189,
+      "step": 1500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 2.5915032679738562e-05,
+      "loss": 1.447,
+      "step": 2000
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.48937908496732e-05,
+      "loss": 1.4006,
+      "step": 2500
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 2.3872549019607843e-05,
+      "loss": 1.35,
+      "step": 3000
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.2851307189542485e-05,
+      "loss": 1.3541,
+      "step": 3500
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 2.1830065359477124e-05,
+      "loss": 1.0592,
+      "step": 4000
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 2.0808823529411766e-05,
+      "loss": 0.9374,
+      "step": 4500
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.9787581699346405e-05,
+      "loss": 0.9286,
+      "step": 5000
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.8766339869281047e-05,
+      "loss": 0.9295,
+      "step": 5500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.7745098039215686e-05,
+      "loss": 0.9148,
+      "step": 6000
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.6723856209150328e-05,
+      "loss": 0.9295,
+      "step": 6500
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.5702614379084967e-05,
+      "loss": 0.8957,
+      "step": 7000
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 1.4681372549019607e-05,
+      "loss": 0.7651,
+      "step": 7500
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 1.366013071895425e-05,
+      "loss": 0.4515,
+      "step": 8000
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 1.2638888888888888e-05,
+      "loss": 0.4702,
+      "step": 8500
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.161764705882353e-05,
+      "loss": 0.4578,
+      "step": 9000
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 1.059640522875817e-05,
+      "loss": 0.4478,
+      "step": 9500
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 9.575163398692811e-06,
+      "loss": 0.4501,
+      "step": 10000
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 8.553921568627451e-06,
+      "loss": 0.4619,
+      "step": 10500
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 7.532679738562092e-06,
+      "loss": 0.4468,
+      "step": 11000
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 6.511437908496732e-06,
+      "loss": 0.1485,
+      "step": 11500
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 5.490196078431373e-06,
+      "loss": 0.1502,
+      "step": 12000
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 4.468954248366013e-06,
+      "loss": 0.1412,
+      "step": 12500
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 3.447712418300654e-06,
+      "loss": 0.1327,
+      "step": 13000
+    },
+    {
+      "epoch": 3.68,
+      "learning_rate": 2.4264705882352943e-06,
+      "loss": 0.1521,
+      "step": 13500
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 1.4052287581699345e-06,
+      "loss": 0.1475,
+      "step": 14000
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 3.839869281045752e-07,
+      "loss": 0.1381,
+      "step": 14500
+    },
+    {
+      "epoch": 4.0,
+      "step": 14688,
+      "total_flos": 3.682861839160443e+17,
+      "train_loss": 0.7583477986664034,
+      "train_runtime": 15779.9103,
+      "train_samples_per_second": 33.507,
+      "train_steps_per_second": 0.931
+    }
+  ],
+  "max_steps": 14688,
+  "num_train_epochs": 4,
+  "total_flos": 3.682861839160443e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f8d74524607e54cdde919fd1ee8d4ad432debb904f9e46b7d947af823703b3e
+size 2735