Initial commit

README.md

@@ -0,0 +1,75 @@
+---
+tags:
+- generated_from_trainer
+model-index:
+- name: aristoBERTo
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# aristoBERTo
+
+This model is a fine-tuned version of [nlpaueb/bert-base-greek-uncased-v1](https://huggingface.co/nlpaueb/bert-base-greek-uncased-v1) on the Ancient Greek dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.6323
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 20.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step    | Validation Loss |
+|:-------------:|:-----:|:-------:|:---------------:|
+| 2.2017        | 1.0   | 170711  | 2.4411          |
+| 2.0302        | 2.0   | 341422  | 2.2853          |
+| 1.9459        | 3.0   | 512133  | 2.1911          |
+| 1.8847        | 4.0   | 682844  | 2.1187          |
+| 1.85          | 5.0   | 853555  | 2.0841          |
+| 1.834         | 6.0   | 1024266 | 2.0267          |
+| 1.7599        | 7.0   | 1194977 | 1.9874          |
+| 1.7147        | 8.0   | 1365688 | 1.9495          |
+| 1.6882        | 9.0   | 1536399 | 1.9030          |
+| 1.6433        | 10.0  | 1707110 | 1.8845          |
+| 1.6289        | 11.0  | 1877821 | 1.8485          |
+| 1.5758        | 12.0  | 2048532 | 1.8097          |
+| 1.5439        | 13.0  | 2219243 | 1.7916          |
+| 1.5358        | 14.0  | 2389954 | 1.7680          |
+| 1.4946        | 15.0  | 2560665 | 1.7360          |
+| 1.4753        | 16.0  | 2731376 | 1.7052          |
+| 1.4628        | 17.0  | 2902087 | 1.6804          |
+| 1.4223        | 18.0  | 3072798 | 1.6720          |
+| 1.3909        | 19.0  | 3243509 | 1.6521          |
+| 1.377         | 20.0  | 3414220 | 1.6314          |
+
+
+### Framework versions
+
+- Transformers 4.14.0.dev0
+- Pytorch 1.10.0+cu102
+- Datasets 1.16.1
+- Tokenizers 0.10.3

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "nlpaueb/bert-base-greek-uncased-v1",
+  "architectures": [
+    "BertForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.14.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 35000
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 20.0,
+    "eval_loss": 1.6322814226150513,
+    "eval_runtime": 77.9898,
+    "eval_samples": 84476,
+    "eval_samples_per_second": 1083.168,
+    "eval_steps_per_second": 67.701,
+    "perplexity": 5.1155321064018535
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eae6b4506f643cf06b746b229983045ebe08ff7ba45fe3ce0d657914140c2d04
+size 451916331

special_tokens_map.json

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

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": "/tmp/xdg-cache/huggingface/transformers/347c648cda48ba9d519777e5e1e677a659862faeb5ad9e219ff0a95c62cfdea4.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d", "name_or_path": "nlpaueb/bert-base-greek-uncased-v1", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}