Update README.md

README.md

@@ -1,71 +1,115 @@
 ---
-library_name: transformers
+'[object Object]': null
 license: apache-2.0
-base_model: distilbert-base-uncased
-tags:
-- generated_from_trainer
+datasets:
+- maryzhang/hw1-24679-image-dataset
+language:
+- en
 metrics:
 - accuracy
-- f1
-- precision
-- recall
-model-index:
-- name: finetuned_model
-  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. -->
+# Model Card for {{ model_id | default("Model ID", true) }}
 
-# finetuned_model
+<!-- Provide a quick summary of what the model is/does. -->
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the james-kramer/football_news dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.0000
-- Accuracy: 1.0
-- F1: 1.0
-- Precision: 1.0
-- Recall: 1.0
+This is finetuned version of DistilBERT that is used for sentiment analysis on NFL news titles. 
 
-## Model description
+## Model Details
 
-More information needed
+### Model Description
 
-## Intended uses & limitations
+This model uses the DistilBERT model to classify NFL news article titles as positive or negative. 
 
-More information needed
+- **Developed by:** Devin DeCosmo
+- **Model type:** Binary Sentiment Analysis
+- **Language(s) (NLP):** English
+- **License:** MIT
+- **Finetuned from model:** DistilBERT
 
-## Training and evaluation data
+## Uses
 
-More information needed
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
 
-## Training procedure
+This is used for sentiment analysis of NFL articles, but could possibly be used for other article titles. 
 
-### Training hyperparameters
+### Direct Use
 
-The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 8
-- eval_batch_size: 8
-- seed: 42
-- optimizer: Use OptimizerNames.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: 5
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
 
-### Training results
+The direct use is to classify NFL articles as positive or negative.
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
-| 0.0002        | 1.0   | 80   | 0.0710          | 0.9875   | 0.9875 | 0.9878    | 0.9875 |
-| 0.0001        | 2.0   | 160  | 0.0001          | 1.0      | 1.0    | 1.0       | 1.0    |
-| 0.0001        | 3.0   | 240  | 0.0001          | 1.0      | 1.0    | 1.0       | 1.0    |
-| 0.0001        | 4.0   | 320  | 0.0000          | 1.0      | 1.0    | 1.0       | 1.0    |
-| 0.0001        | 5.0   | 400  | 0.0000          | 1.0      | 1.0    | 1.0       | 1.0    |
+### Out-of-Scope Use
 
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
 
-### Framework versions
+If the dataset was expanded, this could be used for sentiment analysis on other types of articles or find other features like bias towards a team or player.
 
-- Transformers 4.56.1
-- Pytorch 2.8.0+cu126
-- Datasets 4.0.0
-- Tokenizers 0.22.0
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+This is trained off a small dataset of 100 titles, this small dataset could be liable to overfitting and is not robust. 
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+The small dataset size means this model is not highly generalizable. 
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+James-kramer/football_news
+
+This is the training dataset used. 
+It consists of 100 original titles used for validation along with 1000 synthetic pieces of data from training. 
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+This model was trained with DistilBERT using binary classification, a training split of 80%, and 5 epochs.
+I initially used more but this converged extremely quickly. 
+
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+James-kramer/football_news
+The testing data was the 'original' split, the 100 original titles in this set. 
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+This dataset is evaluating whether the food is positive, "1", or negative, "0".
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+The testing metric used was accuracy to ensure the highest accuracy of the model possible. 
+I also considered testing time. This small langauge model ran extremely quickly with 102 steps per second. 
+
+### Results
+
+After training with the initial dataset, this model reached an accuracy of 100% in validation. 
+This is likely due to the simplicity of the task, binary classification, along with distilBERT being made for tasks such as this.
+
+#### Summary
+
+This model reached a high accuracy with our current model, but this perfomance can not be confirmed to continue as the dataset was very small.
+Additional testing with more samples would be highly beneficial.