model update

README.md

@@ -0,0 +1,158 @@
+
+---
+license: cc-by-4.0
+metrics:
+- bleu4
+- meteor
+- rouge-l
+- bertscore
+- moverscore
+language: de
+datasets:
+- lmqg/qag_dequad
+pipeline_tag: text2text-generation
+tags:
+- questions and answers generation
+widget:
+- text: "Empfangs- und Sendeantenne sollen in ihrer Polarisation übereinstimmen, andernfalls wird die Signalübertragung stark gedämpft. "
+  example_title: "Questions & Answers Generation Example 1" 
+model-index:
+- name: lmqg/mt5-base-dequad-qag
+  results:
+  - task:
+      name: Text2text Generation
+      type: text2text-generation
+    dataset:
+      name: lmqg/qag_dequad
+      type: default
+      args: default
+    metrics:
+    - name: BLEU4 (Question & Answer Generation)
+      type: bleu4_question_answer_generation
+      value: 0.0
+    - name: ROUGE-L (Question & Answer Generation)
+      type: rouge_l_question_answer_generation
+      value: 7.56
+    - name: METEOR (Question & Answer Generation)
+      type: meteor_question_answer_generation
+      value: 5.02
+    - name: BERTScore (Question & Answer Generation)
+      type: bertscore_question_answer_generation
+      value: 53.64
+    - name: MoverScore (Question & Answer Generation)
+      type: moverscore_question_answer_generation
+      value: 46.71
+    - name: QAAlignedF1Score-BERTScore (Question & Answer Generation)
+      type: qa_aligned_f1_score_bertscore_question_answer_generation
+      value: 0.1
+    - name: QAAlignedRecall-BERTScore (Question & Answer Generation)
+      type: qa_aligned_recall_bertscore_question_answer_generation
+      value: 0.1
+    - name: QAAlignedPrecision-BERTScore (Question & Answer Generation)
+      type: qa_aligned_precision_bertscore_question_answer_generation
+      value: 0.1
+    - name: QAAlignedF1Score-MoverScore (Question & Answer Generation)
+      type: qa_aligned_f1_score_moverscore_question_answer_generation
+      value: 0.1
+    - name: QAAlignedRecall-MoverScore (Question & Answer Generation)
+      type: qa_aligned_recall_moverscore_question_answer_generation
+      value: 0.1
+    - name: QAAlignedPrecision-MoverScore (Question & Answer Generation)
+      type: qa_aligned_precision_moverscore_question_answer_generation
+      value: 0.1
+---
+
+# Model Card of `lmqg/mt5-base-dequad-qag`
+This model is fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) for question & answer pair generation task on the [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation).
+
+
+### Overview
+- **Language model:** [google/mt5-base](https://huggingface.co/google/mt5-base)   
+- **Language:** de  
+- **Training data:** [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) (default)
+- **Online Demo:** [https://autoqg.net/](https://autoqg.net/)
+- **Repository:** [https://github.com/asahi417/lm-question-generation](https://github.com/asahi417/lm-question-generation)
+- **Paper:** [https://arxiv.org/abs/2210.03992](https://arxiv.org/abs/2210.03992)
+
+### Usage
+- With [`lmqg`](https://github.com/asahi417/lm-question-generation#lmqg-language-model-for-question-generation-)
+```python
+from lmqg import TransformersQG
+
+# initialize model
+model = TransformersQG(language="de", model="lmqg/mt5-base-dequad-qag")
+
+# model prediction
+question_answer_pairs = model.generate_qa("das erste weltweit errichtete Hermann Brehmer 1855 im niederschlesischen ''Görbersdorf'' (heute Sokołowsko, Polen).")
+
+```
+
+- With `transformers`
+```python
+from transformers import pipeline
+
+pipe = pipeline("text2text-generation", "lmqg/mt5-base-dequad-qag")
+output = pipe("Empfangs- und Sendeantenne sollen in ihrer Polarisation übereinstimmen, andernfalls wird die Signalübertragung stark gedämpft. ")
+
+```
+
+## Evaluation
+
+
+- ***Metric (Question & Answer Generation)***: [raw metric file](https://huggingface.co/lmqg/mt5-base-dequad-qag/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qag_dequad.default.json) 
+
+|                                 |   Score | Type    | Dataset                                                            |
+|:--------------------------------|--------:|:--------|:-------------------------------------------------------------------|
+| BERTScore                       |   53.64 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| Bleu_1                          |    4.78 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| Bleu_2                          |    0.75 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| Bleu_3                          |    0.13 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| Bleu_4                          |    0    | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| METEOR                          |    5.02 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| MoverScore                      |   46.71 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| QAAlignedF1Score (BERTScore)    |    0.1  | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| QAAlignedF1Score (MoverScore)   |    0.1  | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| QAAlignedPrecision (BERTScore)  |    0.1  | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| QAAlignedPrecision (MoverScore) |    0.1  | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| QAAlignedRecall (BERTScore)     |    0.1  | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| QAAlignedRecall (MoverScore)    |    0.1  | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+| ROUGE_L                         |    7.56 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
+
+
+
+## Training hyperparameters
+
+The following hyperparameters were used during fine-tuning:
+ - dataset_path: lmqg/qag_dequad
+ - dataset_name: default
+ - input_types: ['paragraph']
+ - output_types: ['questions_answers']
+ - prefix_types: None
+ - model: google/mt5-base
+ - max_length: 512
+ - max_length_output: 256
+ - epoch: 11
+ - batch: 2
+ - lr: 0.0001
+ - fp16: False
+ - random_seed: 1
+ - gradient_accumulation_steps: 32
+ - label_smoothing: 0.15
+
+The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/mt5-base-dequad-qag/raw/main/trainer_config.json).
+
+## Citation
+```
+@inproceedings{ushio-etal-2022-generative,
+    title = "{G}enerative {L}anguage {M}odels for {P}aragraph-{L}evel {Q}uestion {G}eneration",
+    author = "Ushio, Asahi  and
+        Alva-Manchego, Fernando  and
+        Camacho-Collados, Jose",
+    booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
+    month = dec,
+    year = "2022",
+    address = "Abu Dhabi, U.A.E.",
+    publisher = "Association for Computational Linguistics",
+}
+
+```

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "lmqg_output/mt5-base-dequad-qag/best_model",
+  "_name_or_path": "lmqg_output/mt5-base-dequad-qag/model_oprhlh/epoch_5",
   "add_prefix": false,
   "architectures": [
     "MT5ForConditionalGeneration"

eval/metric.first.answer.paragraph.questions_answers.lmqg_qag_dequad.default.json

@@ -0,0 +1 @@
+{"validation": {"Bleu_1": 0.0523300887242649, "Bleu_2": 0.015425329648044056, "Bleu_3": 0.0033981372216489404, "Bleu_4": 0.0011590330614256743, "METEOR": 0.04604401913618217, "ROUGE_L": 0.0653632399467963, "BERTScore": 0.4956596844039635, "MoverScore": 0.4658147968929338, "QAAlignedF1Score (BERTScore)": 0.0004554700964751326, "QAAlignedRecall (BERTScore)": 0.0004551233193738674, "QAAlignedPrecision (BERTScore)": 0.00045581808096671167, "QAAlignedF1Score (MoverScore)": 0.0003364530046369741, "QAAlignedRecall (MoverScore)": 0.0003352767057694357, "QAAlignedPrecision (MoverScore)": 0.000337638268795125}, "test": {"Bleu_1": 0.04784085567656823, "Bleu_2": 0.007466379162325478, "Bleu_3": 0.001282417388919738, "Bleu_4": 7.992728831519406e-08, "METEOR": 0.05021989611520319, "ROUGE_L": 0.07557708033495764, "BERTScore": 0.5363746310588177, "MoverScore": 0.4671249297467373, "QAAlignedF1Score (BERTScore)": 0.0009902584438891626, "QAAlignedRecall (BERTScore)": 0.000979018089808157, "QAAlignedPrecision (BERTScore)": 0.0010017620611794387, "QAAlignedF1Score (MoverScore)": 0.0009763543514696759, "QAAlignedRecall (MoverScore)": 0.0009712798958179098, "QAAlignedPrecision (MoverScore)": 0.0009814842404872709}}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57dbce637b2e3fd0178c455a050f90d5a767c41b1cb4a5eb1dadc0a5da06a684
-size 2329630901
+oid sha256:66e014af24412db5e6d12e2d45c396f55f91d1c9358654ca1d691ac94ff932f8
+size 2329634869

tokenizer_config.json

@@ -2,7 +2,7 @@
   "additional_special_tokens": null,
   "eos_token": "</s>",
   "extra_ids": 0,
-  "name_or_path": "lmqg_output/mt5-base-dequad-qag/best_model",
+  "name_or_path": "lmqg_output/mt5-base-dequad-qag/model_oprhlh/epoch_5",
   "pad_token": "<pad>",
   "sp_model_kwargs": {},
   "special_tokens_map_file": "/home/patrick/.cache/torch/transformers/685ac0ca8568ec593a48b61b0a3c272beee9bc194a3c7241d15dcadb5f875e53.f76030f3ec1b96a8199b2593390c610e76ca8028ef3d24680000619ffb646276",

trainer_config.json

@@ -0,0 +1 @@
+{"dataset_path": "lmqg/qag_dequad", "dataset_name": "default", "input_types": ["paragraph"], "output_types": ["questions_answers"], "prefix_types": null, "model": "google/mt5-base", "max_length": 512, "max_length_output": 256, "epoch": 11, "batch": 2, "lr": 0.0001, "fp16": false, "random_seed": 1, "gradient_accumulation_steps": 32, "label_smoothing": 0.15}