The primary (and most necessary) step of any fine-tuning course of is knowledge assortment. Right here, I extracted title-thumbnail pairs from my channel in a 2-step course of.
First, I used YouTube’s search API to extract the video IDs for all of the movies on my channel. Second, I used YouTube’s video API to extract the title and thumbnail URL of every of my long-form movies (i.e. longer than 3 min).
# imports
from top_secret import my_key
import requests
from isodate import parse_durationimport pandas as pd
import numpy as np
from sentence_transformers import SentenceTransformer
from datasets import DatasetDict, Dataset
channel_id = 'UCa9gErQ9AE5jT2DZLjXBIdA' # my YouTube channel ID
page_token = None # initialize web page token
url = 'https://www.googleapis.com/youtube/v3/search' # YouTube search API # extract video knowledge throughout a number of search outcome pages
video_id_list = []
whereas page_token != 0:
params = {
"key": my_key,
'channelId': channel_id,
'half': ["snippet","id"],
'order': "date",
'maxResults':50,
'pageToken': page_token
}
response = requests.get(url, params=params)
for raw_item in dict(response.json())['items']:
# solely execute for youtube movies
if raw_item['id']['kind'] != "youtube#video":
proceed
# seize video ids
video_id_list.append(raw_item['id']['videoId'])
strive:
# seize subsequent web page token
page_token = dict(response.json())['nextPageToken']
besides:
# if no subsequent web page token kill whereas loop
page_token = 0
Observe that you will want a YouTube API key to run the above Python code, which you’ll be able to create utilizing the Google Cloud Console. To adapt this to your channel, you simply want to alter the channel_id variable.
# extract video titles and thumbnails
url = "https://www.googleapis.com/youtube/v3/movies"
video_data_list = []for video_id in video_id_list:
params = {
"half": ["snippet","contentDetails"],
"id": video_id,
"key": my_key,
}
response = requests.get(url, params=params)
raw_dict = dict(response.json())['items'][0]
# solely course of movies longer than 3 minutes
iso_duration = raw_dict['contentDetails']["duration"]
if parse_duration(iso_duration).total_seconds() < 180:
proceed
# extract video knowledge
video_data = {}
video_data['video_id'] = video_id
video_data['title'] = raw_dict['snippet']['title']
video_data['thumbnail_url'] = raw_dict['snippet']['thumbnails']['high']['url']
# append knowledge to record
video_data_list.append(video_data)
As an extra step, I created unfavorable thumbnail-title pairs. We are able to use these in the course of the coaching course of to not solely information the mannequin with examples of which embedding ought to be shut collectively (i.e. constructive pair), but in addition which embedding ought to be far aside (i.e. unfavorable pairs).
To do that, I computed the similarity between all doable title pairs utilizing the sentence transformer library. Then for every constructive pair, I matched the least comparable title as a unfavorable instance (guaranteeing there have been no duplicates).
# retailer knowledge in dataframe
df = pd.DataFrame(video_data_list)# Load the mannequin
mannequin = SentenceTransformer("all-mpnet-base-v2")
# Encode all titles
embeddings = mannequin.encode(df['title'].to_list())
# compute similarities
similarities = mannequin.similarity(embeddings, embeddings)
# match least JDs least much like constructive match because the unfavorable match
similarities_argsorted = np.argsort(similarities.numpy(), axis=1)
negative_pair_index_list = []
for i in vary(len(similarities)):
# Begin with the smallest similarity index for the present row
j = 0
index = int(similarities_argsorted[i][j])
# Make sure the index is exclusive
whereas index in negative_pair_index_list:
j += 1 # Transfer to the following smallest index
index = int(similarities_argsorted[i][j]) # Fetch subsequent smallest index
negative_pair_index_list.append(index)
# add unfavorable pairs to df
df['title_neg'] = df['title'].iloc[negative_pair_index_list].values
Lastly, I created a train-valid-test break up and pushed the dataset to the Hugging Face Hub.
# Shuffle the dataset
df = df.pattern(frac=1, random_state=42).reset_index(drop=True)# Cut up into prepare, validation, and take a look at units
train_frac = 0.7
valid_frac = 0.15
test_frac = 0.15
# outline prepare and validation dimension
train_size = int(train_frac * len(df))
valid_size = int(valid_frac * len(df))
# create prepare, validation, and take a look at datasets
df_train = df[:train_size]
df_valid = df[train_size:train_size + valid_size]
df_test = df[train_size + valid_size:]
# Convert the pandas DataFrames again to Hugging Face Datasets
train_ds = Dataset.from_pandas(df_train)
valid_ds = Dataset.from_pandas(df_valid)
test_ds = Dataset.from_pandas(df_test)
# Mix right into a DatasetDict
dataset_dict = DatasetDict({
'prepare': train_ds,
'legitimate': valid_ds,
'take a look at': test_ds
})
# push knowledge to hub
dataset_dict.push_to_hub("shawhin/yt-title-thumbnail-pairs")
Though we’ve all the info we want for fine-tuning, it’s nonetheless not an appropriate format for coaching. Extra particularly, we have to convert our picture URLs to PIL picture objects and arrange our knowledge into (anchor, constructive, unfavorable) triplets, i.e., a thumbnail, its corresponding title, and unfavorable title, respectively.
We are able to course of all three knowledge splits (i.e. prepare, legitimate, and take a look at) within the following means utilizing the Hugging Face Datasets library.
from PIL import Picture# load dataset
dataset = load_dataset("shawhin/yt-title-thumbnail-pairs")
# outline preprocessing operate
def preprocess(batch):
"""
Preprocessing knowledge with out augmentations for take a look at set
"""
# get photographs from urls
image_list = [Image.open(requests.get(url, stream=True).raw)
for url in batch["thumbnail_url"]]
# return columns with customary names
return {
"anchor": image_list,
"constructive": batch["title"],
"unfavorable": batch["title_neg"]
}
# take away columns not related to coaching
columns_to_remove = [col for col in dataset['train'].column_names
if col not in ['anchor', 'positive', 'negative']]
# apply transformations
dataset = dataset.map(preprocess, batched=True,
remove_columns=columns_to_remove)
It’s necessary that we order our columns as (anchor, constructive, unfavorable) triplets as a result of that is the format anticipated by the loss operate we’ll use throughout coaching (which I realized the exhausting means).
Coaching includes optimizing a mannequin’s parameters to attenuate a loss operate. Nevertheless, this worth (i.e. a contrastive loss) isn’t useful in assessing the mannequin’s efficiency on a downstream activity (e.g. matching titles to thumbnails).
A amount that’s extra insightful, on this case, is the mannequin’s potential to accurately match a given thumbnail to the right title amongst a number of candidates. That is denoted Recall@1.
We are able to implement an evaluator appropriate with the Sentence Transformers library to compute this metric. Because the code is kind of lengthy, I gained’t paste it right here, however the curious reader can discover it in Cell 12 of this notebook.
# operate to create new evaluator given knowledge break up
def create_recall_evaluator(set_name, okay=1):
"""
Create triplet evaluator for "prepare", "legitimate", or "take a look at" break up
"""return ImageTextRetrievalEvaluator(
photographs=dataset[f"{set_name}"]["anchor"],
texts=dataset[f"{set_name}"]["positive"],
title=f"yt-title-thumbnail-{set_name}",
okay=okay
)
# Create new evaluator with Recall@okay
evaluator_recall_train = create_recall_evaluator("prepare", okay=1)
evaluator_recall_valid = create_recall_evaluator("legitimate", okay=1)
print("Practice:", evaluator_recall_train(mannequin))
print("Legitimate:", evaluator_recall_valid(mannequin))
# >> Practice: {'yt-title-thumbnail-train_Recall@1': 0.660377358490566}
# >> Legitimate: {'yt-title-thumbnail-valid_Recall@1': 0.6363636363636364}
We are able to see the mannequin already has first rate efficiency out-of-the-box, with appropriate titles being matched 66% of the time.
There are 3 key issues we should do earlier than coaching the mannequin. Particularly, select which parameters to coach, choose a loss operate, and set hyperparameters.
Trainable Parameters
The important thing limitation of this undertaking is that I’ve solely posted 76 YouTube movies (as of penning this). With the validation and take a look at splits, this leaves solely 53 examples for coaching.
Since we’ve so few coaching examples, limiting the variety of parameters we prepare is a good suggestion. On this case, I solely prepare the ultimate projection layer of the mannequin, which maps the textual content and picture embeddings right into a shared vector area. That is about 1M parameters complete.
# import mannequin
from sentence_transformers import SentenceTransformer
mannequin = SentenceTransformer("sentence-transformers/clip-ViT-L-14")# choose particular layers to coach (word: you'll be able to add extra layers to this record)
trainable_layers_list = ['projection']
# Apply freezing configuration
for title, param in mannequin.named_parameters():
# freeze all params
param.requires_grad = False
# unfreeze layers in trainable_layers_list
if any(layer in title for layer in trainable_layers_list):
param.requires_grad = True
# Depend complete and trainable parameters
total_params = sum(p.numel() for p in mannequin.parameters())
trainable_params = sum(p.numel() for p in mannequin.parameters() if p.requires_grad)print(f"Complete parameters: {total_params:,}")
print(f"Trainable parameters: {trainable_params:,}")
print(f"% of trainable parameters: {100*trainable_params/total_params:.2f}%")
# >> Complete parameters: 427,616,513
# >> Trainable parameters: 1,376,256
# >> % of trainable parameters: 0.32%
Loss operate
Right here, I take advantage of the Multiple Negatives Ranking Loss from the Sentence Transformers library (which works with single negatives like on this case). It really works by maximizing the similarity between constructive pairs whereas minimizing the similarity between unfavorable pairs. Right here’s what the loss operate appears to be like like for the one unfavorable case [2].
from sentence_transformers.losses import MultipleNegativesRankingLoss# outline loss
loss = MultipleNegativesRankingLoss(mannequin)
Hyperparameters
For hyperparameters, I experimented with a handful of decisions manually and picked the selection with the most effective validation loss and Recall@1 efficiency. Listed here are the ultimate decisions.
from sentence_transformers import SentenceTransformerTrainingArguments# hyperparameters
num_epochs = 2
batch_size = 16
lr = 1e-4
finetuned_model_name = "clip-title-thumbnail-embeddings"
train_args = SentenceTransformerTrainingArguments(
output_dir=f"fashions/{finetuned_model_name}",
num_train_epochs=num_epochs,
per_device_train_batch_size=batch_size,
per_device_eval_batch_size=batch_size,
learning_rate=lr,
# Analysis settings
eval_strategy="epoch",
eval_steps=1,
logging_steps=1,
)
With our loss and hyperparameters outlined, we will prepare the mannequin utilizing the SentenceTransformersTrainer().
from sentence_transformers import SentenceTransformerTrainercoach = SentenceTransformerTrainer(
mannequin=mannequin,
args=train_args,
train_dataset=dataset["train"],
eval_dataset=dataset["valid"],
loss=loss,
evaluator=[evaluator_recall_train, evaluator_recall_valid],
)
coach.prepare()
Mannequin coaching is an iterative course of the place you could discover dozens of fashions for various decisions of trainable parameters, loss features, and hyperparameters.
Nevertheless, I extremely advocate protecting these experiments so simple as doable. If you end up spending an excessive amount of time tweaking coaching args to get your mannequin to converge, there’s most likely one thing basically unsuitable along with your knowledge (talking from expertise 
).
As a ultimate step, we will consider the mannequin’s Recall@1 rating on the testing set. These knowledge weren’t used for coaching or hyperparameter tuning, so it offers us an unbiased evaluation of the mannequin.
evaluator_recall_test = create_recall_evaluator("take a look at")print("Practice:", evaluator_recall_train(mannequin))
print("Legitimate:", evaluator_recall_valid(mannequin))
print("Take a look at:", evaluator_recall_test(mannequin))
# >> Practice: {'yt-title-thumbnail-train_Recall@1': 0.8490566037735849}
# >> Legitimate: {'yt-title-thumbnail-valid_Recall@1': 0.9090909090909091}
# >> Take a look at: {'yt-title-thumbnail-test_Recall@1': 0.75}
We see that the mannequin performs nicely throughout all three datasets with 75% Recall@1 on the take a look at set. In different phrases, 75% of the time, the mannequin accurately matches a given thumbnail to its authentic title. Moreover, the recall for the validation dataset will increase by 27%!
Multimodal embedding fashions, like CLIP, unlock numerous 0-shot use circumstances resembling picture classification and retrieval. Right here, we noticed how we will fine-tune such a mannequin to adapt it to a specialised area (i.e. my YouTube titles and thumbnails).
Though CLIP is a small mannequin by as we speak’s requirements (~500M parameters) and our coaching dataset was tiny, the ultimate mannequin nonetheless demonstrated robust efficiency on this activity. This highlights the facility of fine-tuning.
When you have any questions or ideas for future content material, let me know within the feedback 
Extra on Multimodal AI 
