Update for requirement instalation

AMQ.py

@@ -35,10 +35,10 @@ class TextandImageQuery():
         Apply an NN to answer the question
         '''
         raw_image=Image.open(raw_image_path).convert("RGB")
-        pipe = pipeline("visual-question-answering", model="Salesforce/blip-vqa-base",device=device)
+        pipe = pipeline("image-text-to-text", model="Salesforce/blip-vqa-base",device=device)
 
         output = pipe(raw_image, question, top_k=1)[0]
-        return output['answer']
+        return output['generated_text']
 
     def run(self):
         self.AnswerText = self.funcTextandImageQuery(self.RawImage, self.QuestionText)

README.md

@@ -20,6 +20,11 @@ The OSD-AMQ Reference Software is found at the NNW gitlab site. It contains:
 Code was designed and tested on an Ubuntu 20.04 operating system using anaconda 23.7.2 and Python 3.9.
 An environment with all the necessary libraries can be created using:
 ```bash
-conda create --name <env> --file requirements.txt
+conda create --name <env>
+pip install -r requirements.txt
+```
+Based on your ffmpeg installation you might also needs :
+```bash
+conda install -c conda-forge ffmpeg
 ```
 

requirements.txt

@@ -3,3 +3,4 @@ transformers
 datasets
 pillow
 soundfile
+sentencepiece
\ No newline at end of file