Model save

.gitattributes

@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
+final_model/tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,160 +1,81 @@
 ---
-language: multilingual
-license: apache-2.0
 tags:
-- sentiment-analysis
-- text-classification
-- xlm-roberta
-- amazon-reviews
-datasets:
-- amazon-reviews
+- generated_from_trainer
 metrics:
 - accuracy
+- precision
+- recall
+- f1
 model-index:
-- name: anpmts/sentiment-classifier
-  results:
-  - task:
-      type: text-classification
-      name: Sentiment Analysis
-    dataset:
-      type: amazon-reviews
-      name: Amazon Reviews
-    metrics:
-    - type: accuracy
-      value: 0.924
-      name: Validation Accuracy
+- name: sentiment-classifier
+  results: []
 ---
 
-# Sentiment Classifier - XLM-RoBERTa
+<!-- 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. -->
 
-This is a sentiment classification model fine-tuned on Amazon Reviews dataset.
+# sentiment-classifier
 
-## Model Description
+This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6947
+- Accuracy: 0.4901
+- Precision: 0.2402
+- Recall: 0.4901
+- F1: 0.3224
+- F1 Macro: 0.3289
+- F1 Negative: 0.0
+- Precision Negative: 0.0
+- Recall Negative: 0.0
+- Support Negative: 900
+- F1 Neutral: 0.6578
+- Precision Neutral: 0.4901
+- Recall Neutral: 1.0
+- Support Neutral: 865
 
-- **Base Model**: xlm-roberta-base
-- **Task**: Sentiment Classification (negative/neutral/positive)
-- **Architecture**: Sequence Classification (single-head)
-- **Languages**: Multilingual (100+ languages)
-- **Parameters**: 278M
+## Model description
 
-## Training Data
+More information needed
 
-- **Dataset**: Amazon Reviews (Kaggle)
-- **Training Samples**: 8,500
-- **Validation Samples**: 1,500
-- **Test Samples**: 5,000
+## Intended uses & limitations
 
-## Performance
+More information needed
 
-| Metric | Value |
-|--------|-------|
-| Validation Accuracy | 92.4% |
-| Training Accuracy | 85.4% |
-| Validation Loss | 0.179 |
+## Training and evaluation data
 
-## Training Details
+More information needed
 
-- **Epochs**: 10
-- **Batch Size**: 16
-- **Learning Rate**: 2e-5
-- **Mixed Precision**: FP16
-- **Optimizer**: AdamW
-- **Scheduler**: Linear Warmup + Cosine Decay
+## Training procedure
 
-## Usage
+### Training hyperparameters
 
-### Option 1: Using AutoModelForSequenceClassification (Recommended)
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 256
+- eval_batch_size: 256
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 10
 
-First, make sure the custom model is registered by installing from this repository:
+### Training results
 
-```python
-# If loading from HuggingFace Hub, you need to install trust_remote_code
-from transformers import AutoTokenizer, AutoModelForSequenceClassification
-import torch
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     | F1 Macro | F1 Negative | Precision Negative | Recall Negative | Support Negative | F1 Neutral | Precision Neutral | Recall Neutral | Support Neutral |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:--------:|:-----------:|:------------------:|:---------------:|:----------------:|:----------:|:-----------------:|:--------------:|:---------------:|
+| 1.1656        | 1.0   | 33   | 0.7228          | 0.5099   | 0.2600    | 0.5099 | 0.3444 | 0.3377   | 0.6754      | 0.5099             | 1.0             | 900              | 0.0        | 0.0               | 0.0            | 865             |
+| 0.8474        | 2.0   | 66   | 0.7003          | 0.4901   | 0.2402    | 0.4901 | 0.3224 | 0.3289   | 0.0         | 0.0                | 0.0             | 900              | 0.6578     | 0.4901            | 1.0            | 865             |
+| 0.8033        | 3.0   | 99   | 0.8336          | 0.4901   | 0.2402    | 0.4901 | 0.3224 | 0.3289   | 0.0         | 0.0                | 0.0             | 900              | 0.6578     | 0.4901            | 1.0            | 865             |
+| 0.7789        | 4.0   | 132  | 0.7006          | 0.5099   | 0.2600    | 0.5099 | 0.3444 | 0.3377   | 0.6754      | 0.5099             | 1.0             | 900              | 0.0        | 0.0               | 0.0            | 865             |
+| 0.7639        | 5.0   | 165  | 0.6940          | 0.4901   | 0.2402    | 0.4901 | 0.3224 | 0.3289   | 0.0         | 0.0                | 0.0             | 900              | 0.6578     | 0.4901            | 1.0            | 865             |
+| 0.7385        | 6.0   | 198  | 0.6946          | 0.4901   | 0.2402    | 0.4901 | 0.3224 | 0.3289   | 0.0         | 0.0                | 0.0             | 900              | 0.6578     | 0.4901            | 1.0            | 865             |
+| 0.7299        | 7.0   | 231  | 0.6961          | 0.4901   | 0.2402    | 0.4901 | 0.3224 | 0.3289   | 0.0         | 0.0                | 0.0             | 900              | 0.6578     | 0.4901            | 1.0            | 865             |
+| 0.7287        | 8.0   | 264  | 0.6943          | 0.4901   | 0.2402    | 0.4901 | 0.3224 | 0.3289   | 0.0         | 0.0                | 0.0             | 900              | 0.6578     | 0.4901            | 1.0            | 865             |
 
-# Load model and tokenizer
-model_name = "anpmts/sentiment-classifier"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForSequenceClassification.from_pretrained(
-    model_name,
-    trust_remote_code=True  # Required for custom models
-)
 
-# Prepare input
-text = "This product is amazing! Highly recommend."
-inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=256)
+### Framework versions
 
-# Get prediction
-with torch.no_grad():
-    outputs = model(**inputs)
-    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
-    sentiment = torch.argmax(predictions, dim=-1)
-
-# Map to label
-labels = ["negative", "neutral", "positive"]
-print(f"Sentiment: {labels[sentiment.item()]}")
-print(f"Confidence: {predictions[0][sentiment].item():.2%}")
-```
-
-### Option 2: Using Pipeline (Easiest)
-
-```python
-from transformers import pipeline
-
-# Load sentiment analysis pipeline
-classifier = pipeline(
-    "text-classification",
-    model="anpmts/sentiment-classifier",
-    trust_remote_code=True
-)
-
-# Predict
-result = classifier("This product is amazing! Highly recommend.")
-print(result)
-# Output: [{'label': 'positive', 'score': 0.96}]
-```
-
-### Option 3: Direct Model Loading
-
-```python
-from transformers import AutoTokenizer
-import torch
-
-# You need to have the model code available locally
-from src.models import SentimentClassifier
-
-model = SentimentClassifier.from_pretrained("anpmts/sentiment-classifier")
-tokenizer = AutoTokenizer.from_pretrained("anpmts/sentiment-classifier")
-
-# Inference
-text = "This product is amazing!"
-inputs = tokenizer(text, return_tensors="pt", max_length=256, truncation=True, padding=True)
-outputs = model(**inputs)
-predictions = torch.softmax(outputs["logits"], dim=-1)
-```
-
-## Training Metrics Over Epochs
-
-| Epoch | Train Loss | Val Loss | Val Acc |
-|-------|-----------|----------|---------|
-| 1     | 0.639     | 0.613    | 49.5%   |
-| 5     | 0.551     | 0.455    | 68.9%   |
-| 10    | 0.270     | 0.179    | 92.4%   |
-
-## Citation
-
-If you use this model, please cite:
-
-```
-@misc{sentiment-classifier-xlm-roberta,
-  author = {TrustShop},
-  title = {Sentiment Classifier - XLM-RoBERTa},
-  year = {2025},
-  publisher = {HuggingFace},
-  url = {https://huggingface.co/anpmts/sentiment-classifier}
-}
-```
-
-## License
-
-Apache 2.0
+- Transformers 4.40.2
+- Pytorch 2.9.0+cu128
+- Datasets 2.18.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,42 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.49008498583569404,
+    "eval_f1": 0.3223752948653044,
+    "eval_f1_macro": 0.3288973384030418,
+    "eval_f1_negative": 0.0,
+    "eval_f1_neutral": 0.6577946768060836,
+    "eval_loss": 0.6946861743927002,
+    "eval_precision": 0.24018329334157243,
+    "eval_precision_negative": 0.0,
+    "eval_precision_neutral": 0.49008498583569404,
+    "eval_recall": 0.49008498583569404,
+    "eval_recall_negative": 0.0,
+    "eval_recall_neutral": 1.0,
+    "eval_runtime": 0.7012,
+    "eval_samples_per_second": 2517.135,
+    "eval_steps_per_second": 9.983,
+    "eval_support_negative": 900,
+    "eval_support_neutral": 865,
+    "test_accuracy": 0.502,
+    "test_f1": 0.33555792276964047,
+    "test_f1_macro": 0.33422103861517977,
+    "test_f1_negative": 0.0,
+    "test_f1_neutral": 0.6684420772303595,
+    "test_loss": 0.6933125257492065,
+    "test_precision": 0.252004,
+    "test_precision_negative": 0.0,
+    "test_precision_neutral": 0.502,
+    "test_recall": 0.502,
+    "test_recall_negative": 0.0,
+    "test_recall_neutral": 1.0,
+    "test_runtime": 0.352,
+    "test_samples_per_second": 2840.874,
+    "test_steps_per_second": 11.363,
+    "test_support_negative": 498,
+    "test_support_neutral": 502,
+    "total_flos": 67710593593440.0,
+    "train_loss": 0.838070989558191,
+    "train_runtime": 136.8755,
+    "train_samples_per_second": 601.642,
+    "train_steps_per_second": 2.411
+}

config.json

@@ -20,6 +20,6 @@
   },
   "model_type": "sentiment-classifier",
   "pretrained_model": "xlm-roberta-base",
-  "torch_dtype": "float32",
+  "torch_dtype": "bfloat16",
   "transformers_version": "4.40.2"
 }

eval_results.json

@@ -0,0 +1,20 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.49008498583569404,
+    "eval_f1": 0.3223752948653044,
+    "eval_f1_macro": 0.3288973384030418,
+    "eval_f1_negative": 0.0,
+    "eval_f1_neutral": 0.6577946768060836,
+    "eval_loss": 0.6946861743927002,
+    "eval_precision": 0.24018329334157243,
+    "eval_precision_negative": 0.0,
+    "eval_precision_neutral": 0.49008498583569404,
+    "eval_recall": 0.49008498583569404,
+    "eval_recall_negative": 0.0,
+    "eval_recall_neutral": 1.0,
+    "eval_runtime": 0.7012,
+    "eval_samples_per_second": 2517.135,
+    "eval_steps_per_second": 9.983,
+    "eval_support_negative": 900,
+    "eval_support_neutral": 865
+}

final_model/config.json

@@ -0,0 +1,25 @@
+{
+  "architectures": [
+    "SentimentClassifier"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_sentiment.SentimentClassifierConfig",
+    "AutoModelForSequenceClassification": "sentiment_classifier.SentimentClassifier"
+  },
+  "dropout": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "negative",
+    "1": "neutral",
+    "2": "positive"
+  },
+  "label2id": {
+    "negative": 0,
+    "neutral": 1,
+    "positive": 2
+  },
+  "model_type": "sentiment-classifier",
+  "pretrained_model": "xlm-roberta-base",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.2"
+}

final_model/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:345950cfc5b7ef657aecf0810dd83f8e29a5b3dc99780869c74d0b2f67b37952
+size 556116300

final_model/special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

final_model/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0091a328b3441d754e481db5a390d7f3b8dabc6016869fd13ba350d23ddc4cd
+size 17082832

final_model/tokenizer_config.json

@@ -0,0 +1,54 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

final_model/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c0d15bb147b64e7c599e439ac472043fc63dd0a41b956d951ef81d1e2239993
+size 5457

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92ecd57cf94586c75fa3535e22d0eb6ba86d72aa62223dbe93c116d58994f6dc
-size 1112208144
+oid sha256:345950cfc5b7ef657aecf0810dd83f8e29a5b3dc99780869c74d0b2f67b37952
+size 556116300

test_results.json

@@ -0,0 +1,20 @@
+{
+    "epoch": 8.0,
+    "test_accuracy": 0.502,
+    "test_f1": 0.33555792276964047,
+    "test_f1_macro": 0.33422103861517977,
+    "test_f1_negative": 0.0,
+    "test_f1_neutral": 0.6684420772303595,
+    "test_loss": 0.6933125257492065,
+    "test_precision": 0.252004,
+    "test_precision_negative": 0.0,
+    "test_precision_neutral": 0.502,
+    "test_recall": 0.502,
+    "test_recall_negative": 0.0,
+    "test_recall_neutral": 1.0,
+    "test_runtime": 0.352,
+    "test_samples_per_second": 2840.874,
+    "test_steps_per_second": 11.363,
+    "test_support_negative": 498,
+    "test_support_neutral": 502
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "total_flos": 67710593593440.0,
+    "train_loss": 0.838070989558191,
+    "train_runtime": 136.8755,
+    "train_samples_per_second": 601.642,
+    "train_steps_per_second": 2.411
+}