Create app.py
Browse files
app.py
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
import datetime
|
| 3 |
+
import torch
|
| 4 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
| 5 |
+
from datasets import load_dataset, Dataset, DatasetDict
|
| 6 |
+
import huggingface_hub
|
| 7 |
+
import pandas as pd
|
| 8 |
+
|
| 9 |
+
# CONFIG
|
| 10 |
+
MODEL_NAME = "distilbert-base-uncased-finetuned-sst-2-english" # replace with your own
|
| 11 |
+
HF_DATASET_REPO = "your-username/your-logging-dataset" # create on HF Hub
|
| 12 |
+
HF_TOKEN = "hf_..." # your Hugging Face token with write access
|
| 13 |
+
|
| 14 |
+
# Load model + tokenizer
|
| 15 |
+
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
|
| 16 |
+
model = AutoModelForSequenceClassification.from_pretrained(MODEL_NAME)
|
| 17 |
+
|
| 18 |
+
# Setup dataset pushing
|
| 19 |
+
huggingface_hub.login(token=HF_TOKEN)
|
| 20 |
+
|
| 21 |
+
# Store session logs
|
| 22 |
+
log_entries = []
|
| 23 |
+
|
| 24 |
+
def infer_and_log(text_input):
|
| 25 |
+
inputs = tokenizer(text_input, return_tensors="pt", truncation=True)
|
| 26 |
+
with torch.no_grad():
|
| 27 |
+
outputs = model(**inputs)
|
| 28 |
+
logits = outputs.logits.tolist()
|
| 29 |
+
predicted = torch.argmax(outputs.logits, dim=-1).item()
|
| 30 |
+
output_label = model.config.id2label[predicted]
|
| 31 |
+
|
| 32 |
+
# Create log entry
|
| 33 |
+
log_entries.append({
|
| 34 |
+
"timestamp": datetime.datetime.now().isoformat(),
|
| 35 |
+
"input": text_input,
|
| 36 |
+
"logits": logits,
|
| 37 |
+
})
|
| 38 |
+
|
| 39 |
+
return output_label
|
| 40 |
+
|
| 41 |
+
def clear_fields():
|
| 42 |
+
return "", ""
|
| 43 |
+
|
| 44 |
+
def save_to_hf():
|
| 45 |
+
if not log_entries:
|
| 46 |
+
return "Nothing to save."
|
| 47 |
+
|
| 48 |
+
dataset = Dataset.from_pandas(pd.DataFrame(log_entries))
|
| 49 |
+
dataset.push_to_hub(HF_DATASET_REPO)
|
| 50 |
+
log_entries.clear()
|
| 51 |
+
return "Data saved to Hugging Face!"
|
| 52 |
+
|
| 53 |
+
with gr.Blocks() as demo:
|
| 54 |
+
gr.Markdown("### 🔤 Text Classification Demo")
|
| 55 |
+
|
| 56 |
+
with gr.Row():
|
| 57 |
+
input_box = gr.Textbox(label="Input Text", lines=5, interactive=True)
|
| 58 |
+
output_box = gr.Textbox(label="Predicted Label", lines=5)
|
| 59 |
+
|
| 60 |
+
with gr.Row():
|
| 61 |
+
submit_btn = gr.Button("Submit")
|
| 62 |
+
clear_btn = gr.Button("Clear")
|
| 63 |
+
|
| 64 |
+
status_box = gr.Textbox(label="Status", interactive=False)
|
| 65 |
+
|
| 66 |
+
submit_btn.click(fn=infer_and_log, inputs=input_box, outputs=output_box)
|
| 67 |
+
clear_btn.click(fn=clear_fields, outputs=[input_box, output_box])
|
| 68 |
+
|
| 69 |
+
gr.Button("Save Logs to Hub").click(fn=save_to_hf, outputs=status_box)
|
| 70 |
+
|
| 71 |
+
if __name__ == "__main__":
|
| 72 |
+
demo.launch()
|