ILC-CNR/gs-Logion

README.md

@@ -0,0 +1,71 @@
+---
+library_name: transformers
+base_model: cabrooks/LOGION-50k_wordpiece
+tags:
+- generated_from_trainer
+model-index:
+- name: gs-Logion
+  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. -->
+
+# gs-Logion
+
+This model is a fine-tuned version of [cabrooks/LOGION-50k_wordpiece](https://huggingface.co/cabrooks/LOGION-50k_wordpiece) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.3943
+- Top1 Acc: 0.5342
+- Top5 Acc: 0.6849
+- Top10 Acc: 0.7671
+- Top20 Acc: 0.7945
+
+## 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: 8
+- seed: 42
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 10
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Top1 Acc | Top5 Acc | Top10 Acc | Top20 Acc |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|:--------:|:---------:|:---------:|
+| 3.6348        | 1.0   | 1972  | 3.1662          | 0.5059   | 0.6647   | 0.7235    | 0.7471    |
+| 3.1196        | 2.0   | 3944  | 2.8723          | 0.4898   | 0.6531   | 0.7075    | 0.7959    |
+| 2.9131        | 3.0   | 5916  | 2.7926          | 0.5      | 0.6644   | 0.7671    | 0.8014    |
+| 2.7745        | 4.0   | 7888  | 2.6731          | 0.5102   | 0.6871   | 0.7755    | 0.8095    |
+| 2.6649        | 5.0   | 9860  | 2.6617          | 0.5660   | 0.6918   | 0.7484    | 0.7862    |
+| 2.5923        | 6.0   | 11832 | 2.5783          | 0.5759   | 0.7089   | 0.7405    | 0.7722    |
+| 2.5206        | 7.0   | 13804 | 2.5101          | 0.6458   | 0.7708   | 0.8264    | 0.8611    |
+| 2.4581        | 8.0   | 15776 | 2.4972          | 0.5874   | 0.7343   | 0.7972    | 0.8252    |
+| 2.4239        | 9.0   | 17748 | 2.4337          | 0.5685   | 0.7329   | 0.7740    | 0.8082    |
+| 2.401         | 10.0  | 19720 | 2.4897          | 0.5804   | 0.7343   | 0.8042    | 0.8462    |
+
+
+### Framework versions
+
+- Transformers 4.51.3
+- Pytorch 2.7.0+cu126
+- Datasets 3.5.1
+- Tokenizers 0.21.1

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 10.0,
+    "eval_loss": 2.3943166732788086,
+    "eval_runtime": 935.4163,
+    "eval_samples_per_second": 3.88,
+    "eval_steps_per_second": 0.485,
+    "eval_top10_acc": 0.7671232876712328,
+    "eval_top1_acc": 0.5342465753424658,
+    "eval_top20_acc": 0.7945205479452054,
+    "eval_top5_acc": 0.684931506849315,
+    "step": 19720,
+    "total_flos": 2.076631553015808e+16,
+    "train_loss": 2.7502908888501523,
+    "train_runtime": 4509.892,
+    "train_samples_per_second": 69.971,
+    "train_steps_per_second": 4.373
+}

config.json

@@ -0,0 +1,24 @@
+{
+  "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,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 50000
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 10.0,
+    "eval_loss": 2.3943166732788086,
+    "eval_runtime": 935.4163,
+    "eval_samples_per_second": 3.88,
+    "eval_steps_per_second": 0.485,
+    "eval_top10_acc": 0.7671232876712328,
+    "eval_top1_acc": 0.5342465753424658,
+    "eval_top20_acc": 0.7945205479452054,
+    "eval_top5_acc": 0.684931506849315
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77d84373add14b8c0d1b64f4f782c44a02775aab469d9bd56990a1dfc9bd26f6
+size 497995232

special_tokens_map.json

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

tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "step": 19720,
+    "total_flos": 2.076631553015808e+16,
+    "train_loss": 2.7502908888501523,
+    "train_runtime": 4509.892,
+    "train_samples_per_second": 69.971,
+    "train_steps_per_second": 4.373
+}

trainer_state.json

@@ -0,0 +1,245 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 19720,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 6.913986682891846,
+      "learning_rate": 4.500507099391481e-05,
+      "loss": 3.6348,
+      "step": 1972
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.166245222091675,
+      "eval_runtime": 259.7532,
+      "eval_samples_per_second": 3.823,
+      "eval_steps_per_second": 0.481,
+      "eval_top10_acc": 0.7235294117647059,
+      "eval_top1_acc": 0.5058823529411764,
+      "eval_top20_acc": 0.7470588235294118,
+      "eval_top5_acc": 0.6647058823529411,
+      "step": 1972
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 6.612165927886963,
+      "learning_rate": 4.000760649087221e-05,
+      "loss": 3.1196,
+      "step": 3944
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.872304916381836,
+      "eval_runtime": 260.1173,
+      "eval_samples_per_second": 3.818,
+      "eval_steps_per_second": 0.481,
+      "eval_top10_acc": 0.7074829931972789,
+      "eval_top1_acc": 0.4897959183673469,
+      "eval_top20_acc": 0.7959183673469388,
+      "eval_top5_acc": 0.6530612244897959,
+      "step": 3944
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 6.386370658874512,
+      "learning_rate": 3.5010141987829614e-05,
+      "loss": 2.9131,
+      "step": 5916
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.7925734519958496,
+      "eval_runtime": 257.3182,
+      "eval_samples_per_second": 3.859,
+      "eval_steps_per_second": 0.486,
+      "eval_top10_acc": 0.7671232876712328,
+      "eval_top1_acc": 0.5,
+      "eval_top20_acc": 0.8013698630136986,
+      "eval_top5_acc": 0.6643835616438356,
+      "step": 5916
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 8.058805465698242,
+      "learning_rate": 3.0010141987829615e-05,
+      "loss": 2.7745,
+      "step": 7888
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.6731116771698,
+      "eval_runtime": 260.3384,
+      "eval_samples_per_second": 3.814,
+      "eval_steps_per_second": 0.48,
+      "eval_top10_acc": 0.7755102040816326,
+      "eval_top1_acc": 0.5102040816326531,
+      "eval_top20_acc": 0.8095238095238095,
+      "eval_top5_acc": 0.6870748299319728,
+      "step": 7888
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 7.619845390319824,
+      "learning_rate": 2.5010141987829615e-05,
+      "loss": 2.6649,
+      "step": 9860
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.6617324352264404,
+      "eval_runtime": 257.8712,
+      "eval_samples_per_second": 3.851,
+      "eval_steps_per_second": 0.485,
+      "eval_top10_acc": 0.7484276729559748,
+      "eval_top1_acc": 0.5660377358490566,
+      "eval_top20_acc": 0.7861635220125787,
+      "eval_top5_acc": 0.6918238993710691,
+      "step": 9860
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 6.427279949188232,
+      "learning_rate": 2.001521298174442e-05,
+      "loss": 2.5923,
+      "step": 11832
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.5783021450042725,
+      "eval_runtime": 257.5407,
+      "eval_samples_per_second": 3.856,
+      "eval_steps_per_second": 0.485,
+      "eval_top10_acc": 0.740506329113924,
+      "eval_top1_acc": 0.5759493670886076,
+      "eval_top20_acc": 0.7721518987341772,
+      "eval_top5_acc": 0.7088607594936709,
+      "step": 11832
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 7.530773639678955,
+      "learning_rate": 1.5017748478701827e-05,
+      "loss": 2.5206,
+      "step": 13804
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.5101499557495117,
+      "eval_runtime": 254.4047,
+      "eval_samples_per_second": 3.903,
+      "eval_steps_per_second": 0.491,
+      "eval_top10_acc": 0.8263888888888888,
+      "eval_top1_acc": 0.6458333333333334,
+      "eval_top20_acc": 0.8611111111111112,
+      "eval_top5_acc": 0.7708333333333334,
+      "step": 13804
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 6.4842047691345215,
+      "learning_rate": 1.0017748478701826e-05,
+      "loss": 2.4581,
+      "step": 15776
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.4972355365753174,
+      "eval_runtime": 256.1582,
+      "eval_samples_per_second": 3.877,
+      "eval_steps_per_second": 0.488,
+      "eval_top10_acc": 0.7972027972027972,
+      "eval_top1_acc": 0.5874125874125874,
+      "eval_top20_acc": 0.8251748251748252,
+      "eval_top5_acc": 0.7342657342657343,
+      "step": 15776
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 7.002655029296875,
+      "learning_rate": 5.017748478701826e-06,
+      "loss": 2.4239,
+      "step": 17748
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.4337377548217773,
+      "eval_runtime": 255.3165,
+      "eval_samples_per_second": 3.889,
+      "eval_steps_per_second": 0.49,
+      "eval_top10_acc": 0.773972602739726,
+      "eval_top1_acc": 0.5684931506849316,
+      "eval_top20_acc": 0.8082191780821918,
+      "eval_top5_acc": 0.7328767123287672,
+      "step": 17748
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 6.361621856689453,
+      "learning_rate": 2.281947261663286e-08,
+      "loss": 2.401,
+      "step": 19720
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.4897375106811523,
+      "eval_runtime": 257.3155,
+      "eval_samples_per_second": 3.859,
+      "eval_steps_per_second": 0.486,
+      "eval_top10_acc": 0.8041958041958042,
+      "eval_top1_acc": 0.5804195804195804,
+      "eval_top20_acc": 0.8461538461538461,
+      "eval_top5_acc": 0.7342657342657343,
+      "step": 19720
+    },
+    {
+      "epoch": 10.0,
+      "step": 19720,
+      "total_flos": 2.076631553015808e+16,
+      "train_loss": 2.7502908888501523,
+      "train_runtime": 4509.892,
+      "train_samples_per_second": 69.971,
+      "train_steps_per_second": 4.373
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.3943166732788086,
+      "eval_runtime": 935.4163,
+      "eval_samples_per_second": 3.88,
+      "eval_steps_per_second": 0.485,
+      "eval_top10_acc": 0.7671232876712328,
+      "eval_top1_acc": 0.5342465753424658,
+      "eval_top20_acc": 0.7945205479452054,
+      "eval_top5_acc": 0.684931506849315,
+      "step": 19720
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 19720,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.076631553015808e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56d990e78a87875a305cb75d68d2e66e4c4d1896a4e05bd3d7dde0c1e1ebd2d3
+size 5649