| # import gradio as gr | |
| # def greet(name): | |
| # return "Hello " + name + "!!" | |
| # iface = gr.Interface(fn=greet, inputs="text", outputs="text") | |
| # iface.launch() | |
| # model = AutoModelForSequenceClassification.from_pretrained("tabibu-ai/mental-health-chatbot") | |
| # write a gradio interface for tabibu-ai/mental-health-chatbot in huggingfacehub | |
| # Path: app.py | |
| import pickle | |
| import numpy as np | |
| import gradio as gr | |
| # install transformers and torch in requirements.txt | |
| from transformers import AutoTokenizer, AutoModelForSequenceClassification | |
| from sklearn.feature_extraction.text import TfidfVectorizer | |
| # tokenizer = AutoTokenizer.from_pretrained("tabibu-ai/mental-health-chatbot") | |
| # tokenizer = AutoTokenizer.from_pretrained("rabiaqayyum/autotrain-mental-health-analysis-752423172") | |
| model = pickle.load(open("model.pkl", "rb")) | |
| def classify_text(inp): | |
| # input_ids = tokenizer.encode(inp, return_tensors='pt') | |
| # output = model.predict(input_ids) | |
| # return output.logits.argmax().item() | |
| # vectorizer = TfidfVectorizer() | |
| # X = vectorizer.fit_transform(inp) | |
| # reshape the input to 2D | |
| # convert the input to a numpy array | |
| # return model.predict( np.array(inp).reshape(1, -1) ) | |
| reshaped = np.array(inp).reshape(1, -1) | |
| return model.predict(reshaped) | |
| # # encode the input text | |
| # encoded_input = tokenizer(text, return_tensors='pt') | |
| # # get the prediction | |
| # output = model(**encoded_input) | |
| # # get the label | |
| # label = output[0].argmax().item() | |
| # # return the label | |
| # return label | |
| iface = gr.Interface(fn=classify_text, inputs="text", outputs="label", | |
| interpretation="default", examples=[ | |
| ["I am feeling depressed"], | |
| ["I am feeling anxious"], | |
| ["I am feeling stressed"], | |
| ["I am feeling sad"], | |
| ]) | |
| iface.launch() | |