shawei3000

I tried adding following in the Model sub-class: @tf.function(input_signature=[{"unique_ids": tf.TensorSpec(shape=[ None], dtype=tf.int32), "input_ids": tf.TensorSpec(shape=[ None, max_seq], dtype=tf.int32), "input_mask": tf.TensorSpec(shape=[ None, max_seq], dtype=tf.int32), "segment_ids": tf.TensorSpec(shape=[ None, max_seq], dtype=tf.int32), "cls_index": tf.TensorSpec(shape=[ None],...

Thanks, Yes, in order to reproduce the error: a). add following in class ALBertQAModel(tf.keras.Model), right above "def call(self, inputs):" ------------ @tf.function(input_signature=[{"unique_ids": tf.TensorSpec(shape=[ None], dtype=tf.int32), "input_ids": tf.TensorSpec(shape=[ None, your_max_seq], dtype=tf.int32), "input_mask":...

Hi, @lapolonio : Your code works as below, but I have issue use your serving_input_fun to call service through gRPC client, not likely to pass in example in gRPC, have...

i followed the above steps, and updated newspaper/* files in my specific anaconda env, and still experience significant missing paragraphs for this url, ( https://www.stltoday.com/news/local/crime-and-courts/belleville-man-gets-20-years-for-ponzi-scheme/article_194000a6-1a13-5841-b53a-44305142bd23.html ), maybe this is different...

@renatoviolin , I went through the updated files, could you tell me where/how you freezed the layers: 1~11? Thnx!

let me know if performance increases. with elmo, has potential to reach 84+ (FM)

would this save just fine-tuned Lora in 4 bit, or the merged 4bit whole model? I am looking for a solution to save whole model in 4 bits after fine-tuning...