init

experiments/huggingface_ops.py

@@ -2,8 +2,10 @@ from huggingface_hub import HfApi, ModelFilter
 from pprint import pprint
 
 api = HfApi()
-models = api.list_models(filter=ModelFilter(author='tweettemposhift'))
-models_filtered = [i.modelId for i in models if 'topic-' in i.modelId]
-
-pprint(sorted([i for i in models_filtered if i.endswith('twitter-roberta-base-2019-90m')]))
-pprint(sorted([i for i in models_filtered if i.endswith('twitter-roberta-base-dec2020')]))
+models = api.list_models(filter=ModelFilter(author='vocabtrimmer'))
+models_filtered = [i.modelId for i in models if 'qg-' in i.modelId]
+pprint(sorted(models_filtered))
+# models = api.list_models(filter=ModelFilter(author='tweettemposhift'))
+# models_filtered = [i.modelId for i in models if 'topic-' in i.modelId]
+# pprint(sorted([i for i in models_filtered if i.endswith('twitter-roberta-base-2019-90m')]))
+# pprint(sorted([i for i in models_filtered if i.endswith('twitter-roberta-base-dec2020')]))

experiments/model_finetuning_ner.py

@@ -84,22 +84,6 @@ def main(
 
     tokenizer = AutoTokenizer.from_pretrained(model, add_prefix_space=True, use_fast=True)
 
-    def align_labels_with_tokens(labels, word_ids):
-        new_labels = []
-        current_word = None
-        for word_id in word_ids:
-            if word_id is None:
-                label = -100
-            elif word_id != current_word:
-                current_word = word_id
-                label = labels[word_id]
-            else:
-                label = labels[word_id]
-                if ID2LABEL[label].startswith("B-"):  # If the label is B-XXX we change it to I-XXX
-                    label = LABEL2ID[ID2LABEL[label].replace("B-", "I-")]
-            new_labels.append(label)
-        return new_labels
-
     def tokenize_and_align_labels(examples):
         tokens = [[preprocess(model, w) for w in t] for t in examples["text_tokenized"]]
         tokenized_inputs = tokenizer(
@@ -110,11 +94,6 @@ def main(
             max_length=128)
         all_labels = examples["gold_label_sequence"]
         new_labels = []
-        # if tokenizer.is_fast:
-        #     for ind, labels in enumerate(all_labels):
-        #         word_ids = tokenized_inputs.word_ids(ind)
-        #         new_labels.append(align_labels_with_tokens(labels, word_ids))
-        # else:
         for token, label in zip(tokens, all_labels):
             tmp_labels = [-100]
             for to, la in zip(token, label):
@@ -138,8 +117,8 @@ def main(
         predictions = np.argmax(logits, axis=-1)
         labels_new, predictions_new = [], []
         for label, prediction in zip(labels, predictions):
-            label = [la for la in label if la != -100]
             prediction = [p for la, p in zip(label, prediction) if la != -100]
+            label = [la for la in label if la != -100]
             labels_new.append([ID2LABEL[la] for la in label])
             predictions_new.append([ID2LABEL[p] for p in prediction])
         return seqeval.compute(predictions=predictions_new, references=labels_new)