Initial Commit

README.md

@@ -0,0 +1,78 @@
+---
+license: mit
+base_model: haryoaw/scenario-MDBT-TCR_data-en-cardiff_eng_only
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+- f1
+model-index:
+- name: scenario-KD-PR-CDF-EN-FROM-EN-D2_data-en-cardiff_eng_only55
+  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. -->
+
+# scenario-KD-PR-CDF-EN-FROM-EN-D2_data-en-cardiff_eng_only55
+
+This model is a fine-tuned version of [haryoaw/scenario-MDBT-TCR_data-en-cardiff_eng_only](https://huggingface.co/haryoaw/scenario-MDBT-TCR_data-en-cardiff_eng_only) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.3581
+- Accuracy: 0.4652
+- F1: 0.4628
+
+## 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: 32
+- eval_batch_size: 32
+- seed: 55
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 30
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
+| No log        | 1.72  | 100  | 1.3031          | 0.4771   | 0.4716 |
+| No log        | 3.45  | 200  | 1.3400          | 0.4683   | 0.4652 |
+| No log        | 5.17  | 300  | 1.3825          | 0.4519   | 0.4469 |
+| No log        | 6.9   | 400  | 1.3630          | 0.4506   | 0.4420 |
+| 1.1126        | 8.62  | 500  | 1.3707          | 0.4638   | 0.4582 |
+| 1.1126        | 10.34 | 600  | 1.3829          | 0.4586   | 0.4484 |
+| 1.1126        | 12.07 | 700  | 1.3900          | 0.4515   | 0.4453 |
+| 1.1126        | 13.79 | 800  | 1.3686          | 0.4533   | 0.4524 |
+| 1.1126        | 15.52 | 900  | 1.3663          | 0.4691   | 0.4671 |
+| 0.9617        | 17.24 | 1000 | 1.3568          | 0.4634   | 0.4633 |
+| 0.9617        | 18.97 | 1100 | 1.3790          | 0.4687   | 0.4636 |
+| 0.9617        | 20.69 | 1200 | 1.3537          | 0.4744   | 0.4719 |
+| 0.9617        | 22.41 | 1300 | 1.3759          | 0.4735   | 0.4682 |
+| 0.9617        | 24.14 | 1400 | 1.3573          | 0.4687   | 0.4675 |
+| 0.9417        | 25.86 | 1500 | 1.3581          | 0.4740   | 0.4734 |
+| 0.9417        | 27.59 | 1600 | 1.3547          | 0.4608   | 0.4588 |
+| 0.9417        | 29.31 | 1700 | 1.3581          | 0.4652   | 0.4628 |
+
+
+### Framework versions
+
+- Transformers 4.33.3
+- Pytorch 2.1.1+cu121
+- Datasets 2.14.5
+- Tokenizers 0.13.3

config.json

@@ -0,0 +1,45 @@
+{
+  "_name_or_path": "haryoaw/scenario-MDBT-TCR_data-en-cardiff_eng_only",
+  "architectures": [
+    "DebertaForSequenceClassificationKD"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "layer_norm_eps": 1e-07,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.3",
+  "type_vocab_size": 0,
+  "vocab_size": 251000
+}

eval_results_cardiff.json

@@ -0,0 +1 @@
+{"arabic": {"f1": 0.47453235688897205, "accuracy": 0.4804597701149425, "confusion_matrix": [[194, 71, 25], [136, 107, 47], [87, 86, 117]]}, "english": {"f1": 0.609083980165232, "accuracy": 0.6091954022988506, "confusion_matrix": [[217, 59, 14], [102, 160, 28], [43, 94, 153]]}, "french": {"f1": 0.4350259504383452, "accuracy": 0.4528735632183908, "confusion_matrix": [[139, 132, 19], [59, 190, 41], [78, 147, 65]]}, "german": {"f1": 0.5212412849776652, "accuracy": 0.5206896551724138, "confusion_matrix": [[133, 105, 52], [62, 168, 60], [40, 98, 152]]}, "hindi": {"f1": 0.46299637429447366, "accuracy": 0.46551724137931033, "confusion_matrix": [[107, 97, 86], [64, 145, 81], [49, 88, 153]]}, "italian": {"f1": 0.5351274987638623, "accuracy": 0.5367816091954023, "confusion_matrix": [[182, 62, 46], [61, 150, 79], [72, 83, 135]]}, "portuguese": {"f1": 0.43249807065239493, "accuracy": 0.43333333333333335, "confusion_matrix": [[100, 137, 53], [81, 157, 52], [38, 132, 120]]}, "spanish": {"f1": 0.4963370591959939, "accuracy": 0.503448275862069, "confusion_matrix": [[181, 71, 38], [139, 89, 62], [71, 51, 168]]}, "all": {"f1": 0.5024926884465007, "accuracy": 0.5021551724137931, "confusion_matrix": [[1250, 724, 346], [654, 1205, 461], [507, 773, 1040]]}}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37ff7bf8c6985dad8459c4d7d89e5e6fe88246410200cae3881ea68f977e8442
+size 946740394

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c6a1bf1dc91f5c2abb49763b191a15d1103ea0151e808b2413d1e7ba0434857
+size 4600