aniemore/aniemore

Emotions recognition from audio and text files (only russian language)

aniemore's Issues

Удалить pytorch и torch-like зависимости из requirements.txt как обязательные

Добавление информации в README проекта

Парни, добавьте информацию о том, что необходимо установить ffmpeg как бинарник сам по себе.
Я из-за этого 4 часа своего времени убил, потому что поганый Python показывает мол ФАЙЛ НЕ НАЙДЕН (запись голоса), а не ffmpeg.

Реф: kkroening/ffmpeg-python#153 (comment)

Проблема при инициализации моделей

Некоторые веса в модели инициализируются заново вместо загрузки из HuggingFace. Неизвестно ,как это влияет на результат, но появляется ворнинг.

import torch
from aniemore.recognizers.voice import VoiceRecognizer
from aniemore.models import HuggingFaceModel

model = HuggingFaceModel.Voice.UniSpeech
device = 'cuda' if torch.cuda.is_available() else 'cpu'
vr = VoiceRecognizer(model=model, device=device)

Ниже всплывающие ворнинги

pytorch_model.bin: 100%|███████████████████████████████████████████████████████████| 1.27G/1.27G [03:38<00:00, 5.79MB/s]
Some weights of the model checkpoint at aniemore/unispeech-sat-emotion-russian-resd were not used when initializing UniSpeechSatForSequenceClassification: ['unispeech_sat.encoder.pos_conv_embed.conv.weight_v', 'unispeech_sat.encoder.pos_conv_embed.conv.weight_g']
- This IS expected if you are initializing UniSpeechSatForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
- This IS NOT expected if you are initializing UniSpeechSatForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
Some weights of UniSpeechSatForSequenceClassification were not initialized from the model checkpoint at aniemore/unispeech-sat-emotion-russian-resd and are newly initialized: ['unispeech_sat.encoder.pos_conv_embed.conv.parametrizations.weight.original0', 'unispeech_sat.encoder.pos_conv_embed.conv.parametrizations.weight.original1']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
preprocessor_config.json: 100%|████████████████████████████████████████████████████████| 212/212 [00:00<00:00, 1.48MB/s]
Some weights of the model checkpoint at aniemore/unispeech-sat-emotion-russian-resd were not used when initializing UniSpeechSatForSequenceClassification: ['unispeech_sat.encoder.pos_conv_embed.conv.weight_v', 'unispeech_sat.encoder.pos_conv_embed.conv.weight_g']
- This IS expected if you are initializing UniSpeechSatForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
- This IS NOT expected if you are initializing UniSpeechSatForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
Some weights of UniSpeechSatForSequenceClassification were not initialized from the model checkpoint at aniemore/unispeech-sat-emotion-russian-resd and are newly initialized: ['unispeech_sat.encoder.pos_conv_embed.conv.parametrizations.weight.original0', 'unispeech_sat.encoder.pos_conv_embed.conv.parametrizations.weight.original1']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.