Kishan

icon indicating a website link https://kishan2k2.github.io/Portfolio/

icon location Delhi icon location Trying to improve myself one thing at a time.

Results 10 comments of Kishan

Error in 4 bit quantization in Gemma-2b

Note I have already pip install accelerate, bitsandbytes

How can I make predictions using a pre-trained model?

Load the model, preprocess the text using seq and padding, and then predict the emotion.

every time i do a prediction,the predicted emotion is neutral. i tried with lstm and cnn

Yes getting the same error

normalize() argument 2 must be str, not float

Donot encode the labels i.e. skip the previous line of code encoding = { 'joy': 0, 'sadness': 1, 'fear': 2, 'anger': 3, 'neutral': 4 } # Integer values for each...

Form carry is paid.

They also provide you with email notification of the response received, unlike formcarry which only allows in-app service.

Form carry is paid.

They provide 100 submissions (50 if you dont join their free newsletter). I believe that should be good to go.

[Feature Request]: Linkedin integration

https://www.linkedin.com/help/linkedin/answer/a1341387#:~:text=In%20order%20to%20protect%20our,automate%20activity%20on%20LinkedIn's%20website.

Can I use it for sentimental analysis.

Please let me know if I am wrong. I will first collect a dataset to fine-tune the model. secondly, I will get the embedding of each of the voice audio...

ValueError: The `response.text` quick accessor only works for simple (single-`Part`) text responses. This response is not simple text.Use the `result.parts` accessor or the full `result.candidates[index].content.parts` lookup instead.

> You could try to delete `max_output_tokens` generation_config in the model if you use that It worked for me.

ValueError: The `response.text` quick accessor only works for simple (single-`Part`) text responses. This response is not simple text.Use the `result.parts` accessor or the full `result.candidates[index].content.parts` lookup instead.

> nothing works for me, my content is medical and it considers it unsafe ! anyone found a solution ? I have not tried it yet, but I have heard...