Multimodal Retrieval Augmented Generation (RAG) using the Gemini API in Agent Platform Reviews

Quota does not allow processing of input PDFs. Furthermore one of them has an unprocessable image.

Karl E. · Reviewed over 2 years ago

dataディレクトリ以下のpdfファイルが多すぎて途中で止まる。ファイルを削除して進めせいかTask3が完了しなかった

masakazu i. · Reviewed over 2 years ago

Progress check in Task 3 does not work. I watched a video of another user doing this lab on Youtube. I found out that "Get documents and images from GCS" cell should download 5 files, 'class_a_share.png', 'tac_table_revenue.png', 'data/google-10k-sample-part1.pdf', 'data/google-10k-sample-part2.pdf', and 'google-10k-sample-14pages.pdf'. but the cell here downloads other files irrelevant to the google 10k document and images. I think the contents in 'gs://github-repo/rag/intro_multimodal_rag' changed and that's why the progress check button cannot catch up the progress. I hope this error to be fixed asap.

유빈 박. · Reviewed over 2 years ago

does not work

창우 송. · Reviewed over 2 years ago

Sudhar 2. · Reviewed over 2 years ago

does not work

창우 송. · Reviewed over 2 years ago

Hao N. · Reviewed over 2 years ago

Too many Error

jaehyeong k. · Reviewed over 2 years ago

이석원 Y. · Reviewed over 2 years ago

Couldn't complete. Got an error in Task 4: "FzErrorArgument: code=4: pixmap must be Grayscale, RGB, or CMYK to save as JPEG" on 'data/2022-alphabet-annual-report.pdf'

Duncan D. · Reviewed over 2 years ago

keep getting this error status = StatusCode.RESOURCE_EXHAUSTED details = "Quota exceeded for aiplatform.googleapis.com/online_prediction_requests_per_base_model with base model: textembedding-gecko. Please submit a quota increase request. https://cloud.google.com/vertex-ai/docs/generative-ai/quotas-genai." ------ ----------------------------------------------------- FzErrorArgument: code=4: pixmap must be Grayscale, RGB, or CMYK to save as JPEG

Reza R. · Reviewed over 2 years ago

GSP1231 Task 3. Use the Gemini Pro model The progress check keeps failing. If you look for successful practice videos on YouTube, there are only 5 files downloaded, but in my practice, over 200 files are downloaded.

JW L. · Reviewed over 2 years ago

Lee Y. · Reviewed over 2 years ago

제대로 동작하지 않ㅇ므

MinKyung S. · Reviewed over 2 years ago

형석 김. · Reviewed over 2 years ago

선미 이. · Reviewed over 2 years ago

뽀애앵 G. · Reviewed over 2 years ago

keep getting this error status = StatusCode.RESOURCE_EXHAUSTED details = "Quota exceeded for aiplatform.googleapis.com/online_prediction_requests_per_base_model with base model: textembedding-gecko. Please submit a quota increase request. https://cloud.google.com/vertex-ai/docs/generative-ai/quotas-genai." ------ ----------------------------------------------------- FzErrorArgument: code=4: pixmap must be Grayscale, RGB, or CMYK to save as JPEG

Reza R. · Reviewed over 2 years ago

선미 이. · Reviewed over 2 years ago

good

신용준 S. · Reviewed over 2 years ago

keep getting this error status = StatusCode.RESOURCE_EXHAUSTED details = "Quota exceeded for aiplatform.googleapis.com/online_prediction_requests_per_base_model with base model: textembedding-gecko. Please submit a quota increase request. https://cloud.google.com/vertex-ai/docs/generative-ai/quotas-genai."

Reza R. · Reviewed over 2 years ago

task 3 did not my progress not working

익화 홍. · Reviewed over 2 years ago

효정 윤. · Reviewed over 2 years ago

Specify the PDF folder with multiple PDF 그냥 여기서 계속 오류남 30분씩 기다리고 계속 오류나서 터짐

최명헌 C. · Reviewed over 2 years ago

If I follow the location you gave me """ # Specify the PDF folder with multiple PDF # pdf_folder_path = "/content/data/" # if running in Google Colab/Colab Enterprise pdf_folder_path = "data/" # if running in Vertex AI Workbench. # Specify the image description prompt. Change it image_description_prompt = """Explain what is going on in the image. If it's a table, extract all elements of the table. If it's a graph, explain the findings in the graph. Do not include any numbers that are not mentioned in the image. """ # Extract text and image metadata from the PDF document text_metadata_df, image_metadata_df = get_document_metadata( multimodal_model, # we are passing gemini 1.0 pro vision model pdf_folder_path, image_save_dir="images", image_description_prompt=image_description_prompt, embedding_size=1408, # add_sleep_after_page = True, # Uncomment this if you are running into API quota issues # sleep_time_after_page = 5, # generation_config = # see next cell # safety_settings = # see next cell ) print("\n\n --- Completed processing. ---") """ You cannot go beyond this cell. The same applies if you set generation_config and safety_settings in the next cell. After several attempts, I figured out that I need to set LOCATION="us-central1", delete data/2022-alphabet-annual-report.pdf and proceed. However, you cannot proceed beyond `Task 3. Use the Gemini Pro model` progress. I don't understand why that is. """ In this task, run through the notebook cells to load the model and download the helper functions and get the documents and images from Cloud Storage. """ I simply ran that cell and everything was perfect as written above, but I couldn't get any points for this progress. I have used up all my five opportunities to uncover these things. So I was unable to complete my graduation. Additionally, it takes too long to execute the cell corresponding to # Extract text and image metadata from the PDF document. It seems like this is an area you definitely need to improve.

JeongWoo S. · Reviewed over 2 years ago