---
library_name: peft
base_model: mistralai/Mistral-7B-v0.1
pipeline_tag: text-generation
---
Description: Sentiment detection of COVID-19 tweets\
Original dataset: https://www.kaggle.com/datatattle/covid-19-nlp-text-classification \
---\
Try querying this adapter for free in Lora Land at https://predibase.com/lora-land! \
The adapter_category is Sentiment Detection and the name is Sentiment Detection (COVID-19)\
---\
Sample input: During the COVID-19 pandemic, Twitter users expressed a variety of opinions about government policy, business decisions, and social norms adopted to reduce the spread of the virus. These opinions can be classified into one of five sentiment types.\n\n### The possible sentiment types are: ###\n\n- Extremely Positive\n\n- Positive\n\n- Neutral\n\n- Negative\n\n- Extremely Negative\n\nDetermine the sentiment of the following tweet using the tweet's text, the date and time it was posted, and the location it was posted from. Assign exactly one type of sentiment to the tweet.\n\n### Text: #Panic buying hits #NewYork City as anxious shoppers stock up on food&medical supplies after #healthcare worker in her 30s becomes #BigApple 1st confirmed #coronavirus patient OR a #Bloomberg staged event?



https://t.co/IASiReGPC4



#QAnon #QAnon2018 #QAnon2020 

#Election2020 #CDC https://t.co/29isZOewxu\n\n### Timestamp: 02-03-2020\n\n### Location: Chicagoland\n\n### Sentiment: \
---\
Sample output: Negative\
---\
Try using this adapter yourself!
```
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "mistralai/Mistral-7B-v0.1"
peft_model_id = "predibase/covid"

model = AutoModelForCausalLM.from_pretrained(model_id)
model.load_adapter(peft_model_id)
```