Open aalgirdas opened 1 month ago
I also meet the same problem. my setence: "Sergey dodged the flower pot that Larry threw in disgust." frame: Dodging trigger: "dodged" elements: Dodger="Sergey" Bad_entity="the flower pot that Larry threw in disgust"
but returns no frames!!!! [] please which dalao help me,thanks!!
Take sentence: Mr. Sherlock Holmes , who was usually very late in the mornings , save upon those not infrequent occasions when he was up all night , was seated at the breakfast table . The code :
result = frame_transformer.detect_frames(my_text) returns no frames! result.frames []
have you solve this problem??? Could you help me?
Take sentence: Mr. Sherlock Holmes , who was usually very late in the mornings , save upon those not infrequent occasions when he was up all night , was seated at the breakfast table . The code :
result = frame_transformer.detect_frames(my_text) returns no frames! result.frames []
Hello, I solve the problem. That is because you download a wrong model. I finally find it after a lot of effort.
you should download this model from huggingface. https://huggingface.co/chanind/frame-semantic-transformer-small https://huggingface.co/chanind/frame-semantic-transformer-base
this is the right output.
Results found in: The hallway smelt of boiled cabbage and old rag mats. FRAME: Connecting_architecture Part: hallway FRAME: Give_impression Phenomenon: The hallway Characterization: of boiled cabbage and old rag mats FRAME: Apply_heat Food: cabbage FRAME: Food Food: cabbage FRAME: Age Entity: rag mats
We encountered an issue processing a long sentence on the live demo page (https://chanind.github.io/frame-semantic-transformer/). The sentence, 'Mr. Sherlock Holmes, who was usually very late in the mornings, save upon those not infrequent occasions when he was up all night, was seated at the breakfast table,' appears to be truncated. This suggests that the current implementation might have limitations on sentence length. It's important for the developers to address this limitation to ensure the package can handle a wider range of text lengths.
Take sentence: Mr. Sherlock Holmes , who was usually very late in the mornings , save upon those not infrequent occasions when he was up all night , was seated at the breakfast table . The code :
result = frame_transformer.detect_frames(my_text) returns no frames! result.frames []