My space is stuck on building forever

import gradio as gr
from transformers import pipeline
import os

# Grab the token from the Space’s Secrets
token = os.environ.get(“HF_TOKEN”)

# Load the model directly into the Space’s generous 16GB RAM
print(“Loading model into memory…”)
ner_pipe = pipeline(
“ner”,
model=“AHaldar/MythologicalNER”,
token=token,
aggregation_strategy=“none”
)
print(“Model loaded successfully!”)

def analyze_text(text):
# The pipeline handles tokenization and PyTorch math
results = ner_pipe(text)
# Clean up numpy types for safe JSON transmission
cleaned = []
for r in results:
cleaned.append({
“entity”: str(r.get(“entity_group”, r.get(“entity”, “O”))),
“word”: str(r[“word”]),
“start”: int(r[“start”]),
“end”: int(r[“end”])
})
return cleaned

# Gradio automatically turns this interface into a REST API
demo = gr.Interface(fn=analyze_text, inputs=“text”, outputs=“json”)
demo.launch()

my app.py is stuck on building forever. I do not understand why. My requirements.txt looks something like this:

transformers
torch
1 Like

@AHaldar This is a problem affecting, I think, everyone on the HF platform.

See following thread:

2 Likes

This topic was automatically closed 12 hours after the last reply. New replies are no longer allowed.