Update README.md

README.md

@@ -1,65 +1,56 @@
 ---
+language: en
+license: mit
 library_name: transformers
-base_model: ProsusAI/finbert
+pipeline_tag: text-classification
 tags:
-- generated_from_trainer
+- finbert
+- finance
+- sentiment-analysis
+- tech-stocks
+base_model: ProsusAI/finbert
+datasets:
+- financial_phrasebank
 metrics:
 - accuracy
 - f1
-model-index:
-- name: FinBERT-Tech-Sentiment
-  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. -->
-
 # FinBERT-Tech-Sentiment
 
-This model is a fine-tuned version of [ProsusAI/finbert](https://huggingface.co/ProsusAI/finbert) on an unknown dataset.
-It achieves the following results on the evaluation set:
-- Loss: 2.3548
-- Accuracy: 0.25
-- F1: 0.2639
-
-## Model description
-
-More information needed
-
-## Intended uses & limitations
+This model is a fine-tuned version of [ProsusAI/finbert](https://huggingface.co/ProsusAI/finbert) for sentiment analysis on financial news related to major technology companies.
 
-More information needed
+## Model Description
 
-## Training and evaluation data
+This model was fine-tuned on a filtered subset of the `financial_phrasebank` dataset. Specifically, it was trained on sentences from the `Sentences_50Agree.txt` file that contained keywords for major tech companies (Google, Apple, Microsoft, Amazon, Nvidia, etc.).
 
-More information needed
+Due to the very small size of the filtered dataset (58 samples), this model is intended as a proof-of-concept and its performance is limited.
 
-## Training procedure
+## Evaluation Results
 
-### Training hyperparameters
+The model achieves the following results on the evaluation set:
+* **Loss:** 2.3548
+* **Accuracy:** 0.25
+* **F1 Score:** 0.2639
 
-The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 32
-- eval_batch_size: 32
-- seed: 42
-- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: linear
-- num_epochs: 4
+## How to Use
 
-### Training results
+You can use this model directly with the `pipeline` function from the `transformers` library.
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
-| No log        | 1.0   | 2    | 3.2596          | 0.0833   | 0.0694 |
-| No log        | 2.0   | 4    | 2.7972          | 0.1667   | 0.1282 |
-| No log        | 3.0   | 6    | 2.4866          | 0.1667   | 0.1282 |
-| No log        | 4.0   | 8    | 2.3548          | 0.25     | 0.2639 |
+```python
+from transformers import pipeline
 
+sentiment_analyzer = pipeline(
+    "text-classification",
+    model="esix117/FinBERT-Tech-Sentiment"
+)
 
-### Framework versions
+# Example 1
+result = sentiment_analyzer("Nvidia reported record earnings, beating all estimates.")
+print(result)
+# >> [{'label': 'positive', 'score': 0.8...}]
 
-- Transformers 4.57.1
-- Pytorch 2.9.0+cu130
-- Datasets 4.3.0
-- Tokenizers 0.22.1
+# Example 2
+result = sentiment_analyzer("Apple shares fell after the new product announcement.")
+print(result)
+# >> [{'label': 'negative', 'score': 0.9...}]