fast-bert

fast-bert copied to clipboard

`learner.fit(epochs=1, r=6e-5, validate=True, # Evaluate the model after each epoch schedule_type="warmup_cosine", optimizer_type="lamb")`  Hi, following the official tutorial ("Language Model Fine-tuning) , i get the following error presented in screenshot,...

FirstGalacticEmpire

`args = Box({ "seed": 42, "task_name": 'Medical_language_modelling', "model_name": 'roberta-base', "model_type": 'roberta', "train_batch_size": 16, "learning_rate": 4e-5, "num_train_epochs": 20, "fp16": True, "fp16_opt_level": "O2", "warmup_steps": 1000, "logging_steps": 0, "max_seq_length": 512, "multi_gpu": True if...

FirstGalacticEmpire

learner.fit doesn't show result after every epoch

7

Hi @kaushaltrivedi , I used: ``` learner.fit(epochs=6, lr=6e-5, validate=True. # Evaluate the model after each epoch schedule_type="warmup_cosine") ``` However, that code onlys checks after the whole training, not after each...

miluki01

Classification Metrics usage

13

How can i use the confusion matrix for each class and the other metrics in this link https://github.com/kaushaltrivedi/fast-bert/issues/17 ??

ahmedbahaaeldin

[Suggestion] Pin requirement versions (specifically python-box)

1

Hello, I am the developer of python-box and see that it is a requirement in this repo and has not been version pinned. I suggest that you pin it to...

cdgriffith

MultiLabel Classification ERROR: Target size (torch.Size([176, 1])) must be the same as input size (torch.Size([16, 1]))

7

Hey, The error is that my 11 labels somehow get mapped to 16*11 values for one batch, which does not match the right size of 16. I have set up...

pvester

ValueError: Target size (torch.Size([9, 4])) must be the same as input size (torch.Size([12, 4]))

2

I am doing multi-label classification. Getting this error !

usmansarwar23

can the version pins for tokenizers and transformers be removed, in requirements.txt?

2

I couldn't get installation with the pinned versions of tokenizers and transformers in place, using the latest nightly rustc. Will my fast-bert installation be incorrect without those version pins? Is...

glycerine

Error when using roberta-large-openai-detector for classification

1

Hi , When using roberta-large-openai-detector for multiclass classification , I am getting below error: `RuntimeError: Error(s) in loading state_dict for RobertaForSequenceClassification: size mismatch for classifier.out_proj.weight: copying a param with shape...

saurabh502

I believe I found a bug in the `predict` code for the `BertClassificationPredictor` class. I am loading a trained model like this: ``` self.model = BertClassificationPredictor(model_path=model_dir, label_path=labels_dir, multi_label=True, model_type='distilbert', do_lower_case=False,...

villalbamartin