Add main files

README.md

@@ -1,13 +1,40 @@
 ---
-title: ACC
-emoji: 📚
-colorFrom: red
-colorTo: blue
+title: Accuracy
+emoji: 📈
+colorFrom: blue
+colorTo: red
 sdk: gradio
-sdk_version: 5.40.0
+sdk_version: 4.44.0
 app_file: app.py
 pinned: false
-short_description: Accuracy
+license: apache-2.0
+tags:
+- evaluate
+- metric
+short_description: Accuracy (ACC)
+description: >-
+  The Accuracy (ACC) metric is used to measure the proportion of correctly
+  predicted sequences compared to the total number of sequences. This metric can
+  handle both integer and string inputs by converting them to strings for
+  comparison. The ACC ranges from 0 to 1, where 1 indicates perfect accuracy
+  (all predictions are correct) and 0 indicates complete failure (no predictions
+  are correct). It is particularly useful in tasks such as OCR, digit
+  recognition, sequence prediction, and any task where exact matches are
+  required. The accuracy can be calculated using the formula: ACC = (Number of
+  Correct Predictions) / (Total Number of Predictions) Where a prediction is
+  considered correct if it exactly matches the ground truth sequence after
+  converting both to strings.
 ---
+# Metric Card for ACC
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+## Metric Description
+
+The Accuracy (ACC) metric is used to measure the proportion of correctly predicted sequences compared to the total number of sequences. This metric can handle both integer and string inputs by converting them to strings for comparison.
+The ACC ranges from 0 to 1, where 1 indicates perfect accuracy (all predictions are correct) and 0 indicates complete failure (no predictions are correct).
+It is particularly useful in tasks such as OCR, digit recognition, sequence prediction, and any task where exact matches are required. The accuracy can be calculated using the formula:
+
+ACC = (Number of Correct Predictions) / (Total Number of Predictions)
+  
+  Where:
+
+A prediction is considered correct if it exactly matches the ground truth sequence after converting both to strings.

app.py

@@ -0,0 +1,40 @@
+import evaluate
+import gradio as gr
+
+module = evaluate.load("Bekhouche/ACC")
+
+def compute_acc(dataframe):
+    predictions = dataframe['Predictions'].tolist()
+    references = dataframe['References'].tolist()
+    if len(predictions) != len(references):
+        return "Error: Number of predictions and references must match!"
+    module.add_batch(predictions=predictions, references=references)
+    result = module.compute()
+    return result
+
+def custom_launch_gradio_widget(module):
+    metric_info = module._info()
+    
+    with gr.Blocks() as demo:
+        gr.Markdown(f"### {metric_info.description}")
+        gr.Markdown(f"**Citation:** {metric_info.citation}")
+        gr.Markdown(f"**Inputs Description:** {metric_info.inputs_description}")
+
+        input_data = gr.Dataframe(
+            headers=["Predictions", "References"],
+            row_count=1,
+            label="Input Predictions and References"
+        )
+
+        run_button = gr.Button("Run ACC")
+        output = gr.Textbox(label="ACC Score")
+
+        run_button.click(
+            compute_acc,
+            inputs=input_data,
+            outputs=output,
+        )
+
+    demo.launch()
+
+custom_launch_gradio_widget(module)

requirements.txt

@@ -0,0 +1,2 @@
+evaluate==0.4.3
+gradio==3.50.0