From 9f484de76dde29c22237c8acad21fb27263b79a4 Mon Sep 17 00:00:00 2001 From: zhang Date: Mon, 27 Jun 2022 22:53:39 +0800 Subject: bert model --- .../00_bert_model_architecture_params.ipynb | 858 +++++++++++++++++++++ 1 file changed, 858 insertions(+) create mode 100644 fine_tune/bert/tutorials/00_bert_model_architecture_params.ipynb (limited to 'fine_tune/bert/tutorials') diff --git a/fine_tune/bert/tutorials/00_bert_model_architecture_params.ipynb b/fine_tune/bert/tutorials/00_bert_model_architecture_params.ipynb new file mode 100644 index 0000000..2c36bd2 --- /dev/null +++ b/fine_tune/bert/tutorials/00_bert_model_architecture_params.ipynb @@ -0,0 +1,858 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "ExecuteTime": { + "end_time": "2022-06-27T14:30:07.738410Z", + "start_time": "2022-06-27T14:30:07.735883Z" + } + }, + "outputs": [], + "source": [ + "from transformers import BertModel, BertForSequenceClassification" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "ExecuteTime": { + "end_time": "2022-06-27T14:28:30.235712Z", + "start_time": "2022-06-27T14:28:30.233000Z" + } + }, + "outputs": [], + "source": [ + "model_name = 'bert-base-uncased'" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "ExecuteTime": { + "end_time": "2022-06-27T14:30:35.901753Z", + "start_time": "2022-06-27T14:30:28.225646Z" + } + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertModel: ['cls.predictions.bias', 'cls.seq_relationship.weight', 'cls.predictions.transform.dense.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.decoder.weight', 'cls.predictions.transform.dense.weight']\n", + "- This IS expected if you are initializing BertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n", + "- This IS NOT expected if you are initializing BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n", + "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.bias', 'cls.seq_relationship.weight', 'cls.predictions.transform.dense.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.decoder.weight', 'cls.predictions.transform.dense.weight']\n", + "- This IS expected if you are initializing BertForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n", + "- This IS NOT expected if you are initializing BertForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n", + "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.bias', 'classifier.weight']\n", + "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n" + ] + } + ], + "source": [ + "model = BertModel.from_pretrained(model_name)\n", + "cls_model = BertForSequenceClassification.from_pretrained(model_name)" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "ExecuteTime": { + "end_time": "2022-06-27T14:28:43.695685Z", + "start_time": "2022-06-27T14:28:43.683251Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "BertModel(\n", + " (embeddings): BertEmbeddings(\n", + " (word_embeddings): Embedding(30522, 768, padding_idx=0)\n", + " (position_embeddings): Embedding(512, 768)\n", + " (token_type_embeddings): Embedding(2, 768)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (encoder): BertEncoder(\n", + " (layer): ModuleList(\n", + " (0): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (1): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (2): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (3): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (4): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (5): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (6): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (7): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (8): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (9): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (10): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (11): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " )\n", + " )\n", + " (pooler): BertPooler(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (activation): Tanh()\n", + " )\n", + ")" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "model" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "ExecuteTime": { + "end_time": "2022-06-27T14:30:47.519574Z", + "start_time": "2022-06-27T14:30:47.513082Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "BertForSequenceClassification(\n", + " (bert): BertModel(\n", + " (embeddings): BertEmbeddings(\n", + " (word_embeddings): Embedding(30522, 768, padding_idx=0)\n", + " (position_embeddings): Embedding(512, 768)\n", + " (token_type_embeddings): Embedding(2, 768)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (encoder): BertEncoder(\n", + " (layer): ModuleList(\n", + " (0): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (1): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (2): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (3): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (4): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (5): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (6): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (7): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (8): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (9): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (10): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (11): BertLayer(\n", + " (attention): BertAttention(\n", + " (self): BertSelfAttention(\n", + " (query): Linear(in_features=768, out_features=768, bias=True)\n", + " (key): Linear(in_features=768, out_features=768, bias=True)\n", + " (value): Linear(in_features=768, out_features=768, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " (output): BertSelfOutput(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (intermediate): BertIntermediate(\n", + " (dense): Linear(in_features=768, out_features=3072, bias=True)\n", + " )\n", + " (output): BertOutput(\n", + " (dense): Linear(in_features=3072, out_features=768, bias=True)\n", + " (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " )\n", + " )\n", + " (pooler): BertPooler(\n", + " (dense): Linear(in_features=768, out_features=768, bias=True)\n", + " (activation): Tanh()\n", + " )\n", + " )\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " (classifier): Linear(in_features=768, out_features=2, bias=True)\n", + ")" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cls_model" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### summary\n", + "- bert: encoder of transformer\n", + " - transformer: encoder-decoder(seq2seq)\n", + "- bert\n", + " - embeddings\n", + " - word(token) embedding\n", + " - position embedding\n", + " - token type embedding\n", + " - encoder(12 layer)\n", + " - self attention (kqv)\n", + " - feed forward\n", + " - pooler" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": { + "ExecuteTime": { + "end_time": "2022-06-27T14:44:50.691120Z", + "start_time": "2022-06-27T14:44:50.685668Z" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "embeddings.word_embeddings.weight -> torch.Size([30522, 768]) -> 23440896\n", + "embeddings.position_embeddings.weight -> torch.Size([512, 768]) -> 393216\n", + "embeddings.token_type_embeddings.weight -> torch.Size([2, 768]) -> 1536\n", + "embeddings.LayerNorm.weight -> torch.Size([768]) -> 768\n", + "embeddings.LayerNorm.bias -> torch.Size([768]) -> 768\n", + "encoder.layer.0.attention.self.query.weight -> torch.Size([768, 768]) -> 589824\n" + ] + } + ], + "source": [ + "# 参数量统计\n", + "total_params = 0\n", + "total_learnable_params = 0\n", + "total_embedding_params = 0\n", + "total_encoder_params = 0\n", + "total_pooler_params = 0\n", + "for name, param in model.named_parameters():\n", + " print(name, '->', param.shape, '->', param.numel())\n", + " if 'embedding' in name:\n", + " total_embedding_params += param.numel()\n", + " if 'encoder' in name:\n", + " total_encoder_params += param.numel()\n", + " if 'pooler' in name:\n", + " total_pooler_params += param.numel()\n", + " if param.requires_grad:\n", + " total_learnable_params += param.numel()\n", + " total_params += param.numel()" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": { + "ExecuteTime": { + "end_time": "2022-06-27T14:40:37.188753Z", + "start_time": "2022-06-27T14:40:37.184701Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "109482240" + ] + }, + "execution_count": 15, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "total_params" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "metadata": { + "ExecuteTime": { + "end_time": "2022-06-27T14:40:44.983087Z", + "start_time": "2022-06-27T14:40:44.979805Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "109482240" + ] + }, + "execution_count": 16, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "total_learnable_params" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": { + "ExecuteTime": { + "end_time": "2022-06-27T14:44:19.645024Z", + "start_time": "2022-06-27T14:44:19.641479Z" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "0.21772649152958506\n", + "0.776879099295009\n", + "0.005394409175405983\n" + ] + } + ], + "source": [ + "params = [total_embedding_params, total_encoder_params, total_pooler_params]\n", + "for param in params:\n", + " print(param/sum(params))" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.6.8" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} -- cgit v1.2.3