Build failing due to "ERROR: Command errored out with exit status 1:"

Spaces
1

basically, the title.

I got the app running on my local environment, but when pushed to spaces, the build fails. Ive run 3 apps on my device (harveenchadha/speech2speech, oiTrans, Vakyansh TTS). All return the same error, while running flawlessly on my device.

Test app.py

import torch
import numpy as np
import gradio as gr

from transformers import AutoTokenizer, AutoModelForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained("juliensimon/autonlp-imdb-demo-hf-16622775")
model = AutoModelForSequenceClassification.from_pretrained("juliensimon/autonlp-imdb-demo-hf-16622775")

def predict(review):
    inputs = tokenizer(review, padding=True, truncation=True, return_tensors="pt")
    outputs = model(**inputs)
    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
    predictions = predictions.detach().numpy()[0]
    index = np.argmax(predictions)
    score = predictions[index]
    return "This revied os {:.3f}% {}".format(100*score, "negative" if index == 0 else "positive")

iface = gr.Interface(fn=predict, inputs='text', outputs='text')
iface.launch()

requirements.txt

transformers
torch
2

Can you share a link to your Space or more detailed log with us?

You can see the logs by clicking “See logs” in the menu with your Space’s name next to the “like” button (see screenshot below):

image
image830×132 11.1 KB

3

My version is a clone of german asr comparison.
His version, which was committed about 9 hours before mine, did not have any errors. I also faced a similar issue with the above test app (taken from a YT video), and other spaces that i cloned and ran.

4

Also, looking at recently created spaces, it look like the issue is only faced by users who are using Gradio. Streamlit apps are working. It could be a pip issue

5

Hello @h4d35, issue with pip resolver, rollbacking to old version.
Your space is up.
Sorry for the inconvenience