Training in progress, epoch 1

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "neuralmind/bert-base-portuguese-cased",
+  "_name_or_path": "neuralmind/bert-large-portuguese-cased",
   "architectures": [
     "ToxicityTypeForSequenceClassification"
   ],
@@ -8,13 +8,13 @@
   "directionality": "bidi",
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "hidden_size": 1024,
   "id2label": {
     "0": "NOT-OFFENSIVE",
     "1": "OFFENSIVE"
   },
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "label2id": {
     "NOT-OFFENSIVE": 0,
     "OFFENSIVE": 1
@@ -22,8 +22,8 @@
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
   "model_type": "bert",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
   "output_past": true,
   "pad_token_id": 0,
   "pooler_fc_size": 768,
@@ -33,7 +33,7 @@
   "pooler_type": "first_token_transform",
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
-  "transformers_version": "4.26.0",
+  "transformers_version": "4.26.1",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 29794

last-checkpoint/config.json

@@ -0,0 +1,40 @@
+{
+  "_name_or_path": "neuralmind/bert-large-portuguese-cased",
+  "architectures": [
+    "ToxicityTypeForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "NOT-OFFENSIVE",
+    "1": "OFFENSIVE"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "NOT-OFFENSIVE": 0,
+    "OFFENSIVE": 1
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.26.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 29794
+}

last-checkpoint/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3097373a4812b7efa975fcfffebce910c9be972d188b249ec4aa25c5350ad93
+size 2675416477

last-checkpoint/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45592d7a0658665761f426eada85c30d56ec10d413005c3cec81e85c00eb944b
+size 1337729837

last-checkpoint/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04d6a2651fa34d58dd0137d1ce4c278eb691e0dc0cd07c612660e6723cd2b957
+size 14503

last-checkpoint/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74bb6545357b275e06db2c532a9dc172de544674f98c91ceb7511b4dba5bd269
+size 623

last-checkpoint/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

last-checkpoint/tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "name_or_path": "neuralmind/bert-large-portuguese-cased",
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": "/root/.cache/huggingface/hub/models--neuralmind--bert-large-portuguese-cased/snapshots/aa302f6ea73b759f7df9cad58bd272127b67ec28/special_tokens_map.json",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

last-checkpoint/trainer_state.json

@@ -0,0 +1,40 @@
+{
+  "best_metric": 0.8549217425622928,
+  "best_model_checkpoint": "/opt/ml/checkpoints/toxic-comment-classification-2023-02-26-22-07-06-968/model/checkpoint-1408",
+  "epoch": 1.0,
+  "global_step": 1408,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.217249581414747e-05,
+      "loss": 0.4937,
+      "step": 500
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 3.1787104153700095e-05,
+      "loss": 0.4465,
+      "step": 1000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8546529723000311,
+      "eval_f1": 0.8549217425622928,
+      "eval_loss": 0.41022783517837524,
+      "eval_precision": 0.8668618958335471,
+      "eval_recall": 0.8546529723000311,
+      "eval_runtime": 11.9429,
+      "eval_samples_per_second": 269.03,
+      "eval_steps_per_second": 33.66,
+      "step": 1408
+    }
+  ],
+  "max_steps": 42240,
+  "num_train_epochs": 30,
+  "total_flos": 1552390150961832.0,
+  "trial_name": null,
+  "trial_params": null
+}

last-checkpoint/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb3b71907c1e39b22ef1ab915290602a79a529af398581187aa888ba7af47b38
+size 3695

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5cc09466df4f7451ec73c9e2f59c8ca058f7d217f855947f6d8de3558d655167
-size 435769709
+oid sha256:45592d7a0658665761f426eada85c30d56ec10d413005c3cec81e85c00eb944b
+size 1337729837

tokenizer_config.json

@@ -4,11 +4,11 @@
   "do_lower_case": false,
   "mask_token": "[MASK]",
   "model_max_length": 1000000000000000019884624838656,
-  "name_or_path": "neuralmind/bert-base-portuguese-cased",
+  "name_or_path": "neuralmind/bert-large-portuguese-cased",
   "never_split": null,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
-  "special_tokens_map_file": "/root/.cache/huggingface/hub/models--neuralmind--bert-base-portuguese-cased/snapshots/94d69c95f98f7d5b2a8700c420230ae10def0baa/special_tokens_map.json",
+  "special_tokens_map_file": "/root/.cache/huggingface/hub/models--neuralmind--bert-large-portuguese-cased/snapshots/aa302f6ea73b759f7df9cad58bd272127b67ec28/special_tokens_map.json",
   "strip_accents": null,
   "tokenize_chinese_chars": true,
   "tokenizer_class": "BertTokenizer",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abf0c3eac401447904c24fc61203586d910edc19df73448d1cbb9cc7add0f27e
-size 3567
+oid sha256:eb3b71907c1e39b22ef1ab915290602a79a529af398581187aa888ba7af47b38
+size 3695