Upload 12 files

README.md

@@ -0,0 +1,33 @@
+### Model Summary
+
+This model is a sentiment classification model fine-tuned on top of mBERTu, a state-of-the-art Maltese language model based on multilingual BERT. It is designed to analyze the sentiment of text in the Maltese language and classify it into different sentiment categories.
+
+### Dataset
+
+The model was fine-tuned on a dataset containing Maltese text with sentiment labels. The dataset consists of text samples in the Maltese language, each labeled with one of the following sentiment categories:
+- Positive
+- Neutral
+
+### Model Architecture
+
+The model utilizes the mBERTu architecture, which is a variant of BERT (Bidirectional Encoder Representations from Transformers) specifically optimized for the Maltese language. BERTu is known for its ability to capture contextual information from text and is pre-trained on a large corpus of Maltese text.
+
+### Fine-Tuning
+
+Fine-tuning is the process of adapting a pre-trained model to a specific task, in this case, sentiment classification. The model was fine-tuned on the sentiment-labeled Maltese text dataset using transfer learning. The fine-tuning process involves updating the model's weights and parameters to make it proficient at sentiment analysis.
+
+### Performance
+
+The model's performance can be assessed through various evaluation metrics, including accuracy, precision, recall, and F1-score. It has been fine-tuned to achieve high accuracy in classifying text into the sentiment categories.
+
+### Usage
+
+You can use this model for sentiment analysis of Maltese text. Given a text input, the model can predict whether the sentiment is positive, negative, or neutral. It can be integrated into applications, chatbots, or services to automatically assess the sentiment of user-generated content.
+
+### License
+
+The model is made available under a specific license, and it's important to refer to the terms and conditions of use provided by the model's creator.
+
+### Creator
+
+This fine-tuned sentiment classification model on mBERTu for Maltese is the work of [Daniil Gurgurov]. 

config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "MLRS/mBERTu",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "gradient_checkpointing": false,
+  "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,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.35.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 119547
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:855ee6d29e7371cfdef92bceae93ae41c5b71066d03a66a0a56a314666973291
+size 711443456

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcae910d692681a683d7b648f5c97d10434e8fe891e014d05615407e12e05616
+size 1423007930

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:992edea9e6cdc72c09e6a348a5bc78e90ae15b52eb7c15095de9744c0fee8ebb
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6a1e3232c9dd0558d3b582345b4a537ca2d63d9ae2df7d7189cbdc92c4e384f
+size 1064

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 +1,57 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "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": false,
+  "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]"
+}

trainer_state.json

@@ -0,0 +1,63 @@
+{
+  "best_metric": 1.4215906858444214,
+  "best_model_checkpoint": "./bertu_sentiment_finetuned/checkpoint-1000",
+  "epoch": 200.0,
+  "eval_steps": 1000,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 50.0,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.0306,
+      "step": 500
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 5e-05,
+      "loss": 0.0,
+      "step": 1000
+    },
+    {
+      "epoch": 100.0,
+      "eval_accuracy": 0.8352941176470589,
+      "eval_f1": 0.7666666666666667,
+      "eval_loss": 1.4215906858444214,
+      "eval_runtime": 1.8326,
+      "eval_samples_per_second": 46.383,
+      "eval_steps_per_second": 1.091,
+      "step": 1000
+    },
+    {
+      "epoch": 150.0,
+      "learning_rate": 2.5e-05,
+      "loss": 0.0,
+      "step": 1500
+    },
+    {
+      "epoch": 200.0,
+      "learning_rate": 0.0,
+      "loss": 0.0,
+      "step": 2000
+    },
+    {
+      "epoch": 200.0,
+      "eval_accuracy": 0.8470588235294118,
+      "eval_f1": 0.7796610169491527,
+      "eval_loss": 1.4892604351043701,
+      "eval_runtime": 1.6492,
+      "eval_samples_per_second": 51.542,
+      "eval_steps_per_second": 1.213,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 2000,
+  "num_train_epochs": 200,
+  "save_steps": 1000,
+  "total_flos": 3.131021558784e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bba5e508e8e53a8d73456214b0ff323aa896d83ec3467ffee358edda490febb
+size 4536