Greek Media SBERT upload to hub

1_Pooling/config.json

@@ -0,0 +1,7 @@
+{
+  "word_embedding_dimension": 768,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": true,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false
+}

README.md

@@ -0,0 +1,154 @@
+---
+language:
+  - el
+pipeline_tag: sentence-similarity
+tags:
+- sentence-transformers
+- feature-extraction
+- sentence-similarity
+- transformers
+metrics:
+  - accuracy_cosinus
+  - accuracy_euclidean
+  - accuracy_manhattan
+model-index:
+  - name: st-greek-media-bert-base-uncased
+    results: [
+      {name: STSbenchmark, value: 0.9563965089445283, limit: 0.0, unit: "%", metric: accuracy_cosinus},
+      {name: STSbenchmark, value: 0.9566394253292384, limit: 0.0, unit: "%", metric: accuracy_euclidean},
+      {name: STSbenchmark, value: 0.9565353183072198, limit: 0.0, unit: "%", metric: accuracy_manhattan}
+    ]
+---
+
+# Greek Media SBERT (uncased)
+## Sentence Transformer
+
+This is a [sentence-transformers](https://www.SBERT.net) based on the [Greek Media BERT (uncased)](https://huggingface.co/dimitriz/greek-media-bert-base-uncased) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
+
+<!--- Describe your model here -->
+
+## Usage (Sentence-Transformers)
+
+Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
+
+```
+pip install -U sentence-transformers
+```
+
+Then you can use the model like this:
+
+```python
+from sentence_transformers import SentenceTransformer
+sentences = ["This is an example sentence", "Each sentence is converted"]
+
+model = SentenceTransformer('dimitriz/st-greek-media-bert-base-uncased')
+embeddings = model.encode(sentences)
+print(embeddings)
+```
+
+
+
+## Usage (HuggingFace Transformers)
+Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
+
+```python
+from transformers import AutoTokenizer, AutoModel
+import torch
+
+
+#Mean Pooling - Take attention mask into account for correct averaging
+def mean_pooling(model_output, attention_mask):
+    token_embeddings = model_output[0] #First element of model_output contains all token embeddings
+    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
+    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
+
+
+# Sentences we want sentence embeddings for
+sentences = ['This is an example sentence', 'Each sentence is converted']
+
+# Load model from HuggingFace Hub
+tokenizer = AutoTokenizer.from_pretrained('dimitriz/st-greek-media-bert-base-uncased')
+model = AutoModel.from_pretrained('dimitriz/st-greek-media-bert-base-uncased')
+
+# Tokenize sentences
+encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
+
+# Compute token embeddings
+with torch.no_grad():
+    model_output = model(**encoded_input)
+
+# Perform pooling. In this case, mean pooling.
+sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
+
+print("Sentence embeddings:")
+print(sentence_embeddings)
+```
+
+
+
+## Evaluation Results
+
+<!--- Describe how your model was evaluated -->
+
+For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=dimitriz/st-greek-media-bert-base-uncased)
+
+
+## Training
+The model was trained on a custom dataset containing triplets from the **combined** Greek 'internet', 'social-media' and 'press' domains, described in the paper [DACL](https://...). 
+- The dataset was created by sampling triplets of sentences from the same domain, where the first two sentences are more similar than the third one. 
+- Training objective was to maximize the similarity between the first two sentences and minimize the similarity between the first and the third sentence. 
+- The model was trained for 3 epochs with a batch size of 16 and a maximum sequence length of 512 tokens. 
+- The model was trained on a single NVIDIA RTX A6000 GPU with 48GB of memory.
+
+The model was trained with the parameters:
+
+**DataLoader**:
+
+`torch.utils.data.dataloader.DataLoader` of length 10807 with parameters:
+```
+{'batch_size': 16, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
+```
+
+**Loss**:
+
+`sentence_transformers.losses.TripletLoss.TripletLoss` with parameters:
+  ```
+  {'distance_metric': 'TripletDistanceMetric.EUCLIDEAN', 'triplet_margin': 5}
+  ```
+
+Parameters of the fit()-Method:
+```
+{
+    "epochs": 3,
+    "evaluation_steps": 1000,
+    "evaluator": "sentence_transformers.evaluation.TripletEvaluator.TripletEvaluator",
+    "max_grad_norm": 1,
+    "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
+    "optimizer_params": {
+        "lr": 2e-05
+    },
+    "scheduler": "WarmupLinear",
+    "steps_per_epoch": null,
+    "warmup_steps": 17290,
+    "weight_decay": 0.01
+}
+```
+
+
+## Full Model Architecture
+```
+SentenceTransformer(
+  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
+  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
+)
+```
+
+## Citing & Authors
+
+```@inproceedings{...,
+  title={DACL},
+  author={Zaikis et al.},
+  booktitle={...},
+  year={2023}
+}
+```

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "/home/dimitriz/.cache/torch/sentence_transformers/dimitriz_greek-media-bert-base-uncased",
+  "architectures": [
+    "BertModel"
+  ],
+  "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": "4.28.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 35000
+}

config_sentence_transformers.json

@@ -0,0 +1,7 @@
+{
+  "__version__": {
+    "sentence_transformers": "2.2.2",
+    "transformers": "4.28.0",
+    "pytorch": "2.0.0+cu118"
+  }
+}

eval/triplet_evaluation_results.csv

@@ -0,0 +1,34 @@
+epoch,steps,accuracy_cosinus,accuracy_manhattan,accuracy_euclidean
+0,1000,0.7708083910259746,0.7698714278278069,0.7704960699599188
+0,2000,0.8108375409921399,0.8136310794163066,0.8141342633560633
+0,3000,0.8408897680148526,0.8418961358943661,0.8424166710044593
+0,4000,0.8605312928357017,0.8604792393246925,0.8606874533687298
+0,5000,0.8751930317699929,0.8756268110284039,0.875748269220759
+0,6000,0.8834695400204744,0.8842503426856142,0.8839553727898947
+0,7000,0.8923706904030677,0.8926656602987871,0.8926309579581143
+0,8000,0.8975413391633266,0.897714850866691,0.8978016067183732
+0,9000,0.9009248173789323,0.901983238769455,0.9020526434508007
+0,10000,0.9048461818749675,0.9053146634740513,0.9053493658147242
+0,-1,0.9095483490361425,0.9099994794648899,0.9103812052122916
+1,1000,0.9107976333003661,0.9109190914927212,0.9110405496850763
+1,2000,0.9121857269272813,0.912255131608627,0.9129491784220846
+1,3000,0.9159335797199522,0.9166102753630733,0.9167317335554284
+1,4000,0.9201499141117069,0.9200458070896882,0.9203234258150712
+1,5000,0.9212950913539119,0.9217982752936685,0.9219197334860236
+1,6000,0.9216421147606406,0.9218503288046779,0.9219891381673694
+1,7000,0.9259278538337411,0.9261013655371054,0.9262575260701335
+1,8000,0.9295889507747298,0.9296063019450662,0.9296236531154026
+1,9000,0.9333715059080735,0.9337011781444658,0.9337185293148023
+1,10000,0.9384727499869866,0.9389585827564069,0.9390106362674162
+1,-1,0.9391494456301077,0.9394444155258272,0.9396873319105373
+2,1000,0.9392014991411171,0.9394097131851543,0.9397046830808738
+2,2000,0.942081793416966,0.9421338469279753,0.9422726562906668
+2,3000,0.9455173251435809,0.9454305692918987,0.9456214321655996
+2,4000,0.9474433050509257,0.9475821144136172,0.9477209237763087
+2,5000,0.9499245224090365,0.9500980341124009,0.9503583016674475
+2,6000,0.9523883885968109,0.9523189839154651,0.9525271979595024
+2,7000,0.9531171377509413,0.9530824354102684,0.953342702965315
+2,8000,0.9544358266965107,0.9545225825481929,0.9545919872295386
+2,9000,0.9547307965922301,0.9550778199989589,0.9551298735099683
+2,10000,0.9560494855377996,0.9562403484115004,0.9563097530928462
+2,-1,0.9563965089445283,0.9565353183072198,0.9566394253292384

modules.json

@@ -0,0 +1,14 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  }
+]

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ea4bbca9a18265ab75010ac3747902c628356f065bae71b87ac50ac597566fe
+size 451756589

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 512,
+  "do_lower_case": false
+}

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 +1,15 @@
+{
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}