ResourceExhaustedError while running Document_representation_from_BERT

[MahajanTarun]

I have been trying to run Document_representation_from_BERT on local machine with enough memory, i.e. 8 GB RAM.
All the other TF function runs without this error for other notebooks on my local machine.

But when I load the Patent_BERT model, i.e.
model = tf.compat.v2.saved_model.load(export_dir=MODEL_DIR, tags=['serve']) model = model.signatures['serving_default']

It also gives the similar error at :
docs_embeddings = [] for _, row in df.iterrows(): inputs = get_bert_token_input(row['claims']) response = model(**inputs) avg_embeddings = pooling( tf.reshape(response['encoder_layer'], shape=[1, -1, 1024])) docs_embeddings.append(avg_embeddings.numpy()[0])

Please help me to get this done. I have already spent a lot of time solving the issue, but to no avail.

[twin9458]

I have the same issue with that.
Especially, when I run response = model(**inputs) there is an error.
NameError : name 'response' is not defined.

My code is
model = tf.compat.v2.saved_model.load(export_dir=MODEL_DIR, tags=['serve'])
model = model.signatures['serving_default']
inputs = get_bert_token_input(example_sent)
response = model(**inputs)

Is there any other solution?
I have to solve this... but I can't ....

[twin9458]

Hi, I have

I have been trying to run Document_representation_from_BERT on local machine with enough memory, i.e. 8 GB RAM. All the other TF function runs without this error for other notebooks on my local machine.

But when I load the Patent_BERT model, i.e. model = tf.compat.v2.saved_model.load(export_dir=MODEL_DIR, tags=['serve']) model = model.signatures['serving_default']

It also gives the similar error at : docs_embeddings = [] for _, row in df.iterrows(): inputs = get_bert_token_input(row['claims']) response = model(**inputs) avg_embeddings = pooling( tf.reshape(response['encoder_layer'], shape=[1, -1, 1024])) docs_embeddings.append(avg_embeddings.numpy()[0])

Please help me to get this done. I have already spent a lot of time solving the issue, but to no avail.

Hi, I don't know if I can help, but if you have the same problem as me, try this.

The PatentBERT model uses averagepooling, but I misunderstood the part of averaging.
In my case, I had to find a sentence vector for each input sentence, so I changed it to a list form. You should refer to line 68.
I will also attach information about the data I entered.

I hope this helps 😥
Good day !