This working note documents the participation of CS_Morgan in the ImageCLEFmedical 2024 Caption subtasks, focusing on Caption Prediction and Concept Detection challenges. The primary objectives included training, validating, and testing multimodal Artificial Intelligence (AI) models intended to automate the process of generating captions and identifying multi-concepts of radiology images. The dataset used is a subset of the Radiology Objects in COntext version 2 (ROCOv2) dataset and contains image-caption pairs and corresponding Unified Medical Language System (UMLS) concepts. To address the caption prediction challenge, diferent variants of the Large Language and Vision Assistant (LLaVA) models were experimented with, tailoring them for the medical domain. Additionally, a lightweight Large Multimodal Model (LMM), and MoonDream2, a small Vision Language Model (VLM), were explored. The former is the instruct variant of the Image-aware Decoder Enhanced à la Flamingo with Interleaved Cross-attentionS (IDEFICS) 9B obtained through quantization. Besides LMMs, conventional encoder-decoder models like Vision Generative Pre-trained Transformer 2 (visionGPT2) and Convolutional Neural Network-Transformer (CNN-Transformer) architectures were considered. Consequently, this enabled 10 submissions for the caption prediction task, with the first submission of LLaVA 1.6 on the Mistral 7B weights securing the 2nd position among the participants. This model was adapted using 40.1M parameters and achieved the best performance on the test data across the performance metrics of BERTScore (0.628059), ROUGE (0.250801), BLEU-1 (0.209298), BLEURT (0.317385), METEOR (0.092682), CIDEr (0.245029), and RefCLIPScore (0.815534). For the concept detection task, our single submission based on the ConvMixer architecture-a hybrid approach leveraging CNN and Transformer advantages-ranked 9th with an F1-score of 0.107645. Overall, the evaluations on the test data for the caption prediction task submissions suggest that LMMs, quantized LMMs, and small VLMs, when adapted and selectively fine-tuned using fewer parameters, have ample potential for understanding medical concepts present in images.
The tasks of automatic caption generation and multi-label prediction from medical images have become
crucial for improving healthcare due to the growing availability of medical images from diferent
modalities like X-radiation (X-ray), Computed Tomography (CT), Positron Emission Tomography (PET),
Magnetic Resonance Imaging (MRI), and Ultrasound (US), as well as the significant advancements in
the computing power of modern graphics processing units [
For the ImageCLEFmedical Caption 2024 [
Dataset for both tasks included curated images from ROCOv2 [
LMMs as an extended variation of LLMs mark a major leap forward in AI by handling and comprehending
various data types, including text, images, audio, and video [
During pre-training, the model is initially trained on vast and diverse datasets, enabling it to learn
general representations before being fine-tuned for specific tasks. This involves utilizing large-scale
datasets that include various modalities [
Fine-tuning LMMs involves adjusting all pre-trained model parameters to enhance performance
on specific tasks, such as image captioning. This process is computationally intensive and
resourcedemanding, especially for models with billions of parameters. Despite these challenges, the full
finetuning technique remains popular due to its potential for achieving high accuracy. For instance, models
like BLIP-2 [
As a result, Parameter-Eficient Fine-Tuning (PEFT) [
Figure 2 indicates that the LoRA approach involves two matrices, and . The matrix is the first
step in the adaptation process, projecting high-dimensional input features into a lower-dimensional
latent space. Typically, its shape includes two values: rank and original dimension (e.g., 32 and 4096).
The matrix is the second component, mapping the lower-dimensional features back to the original
high-dimensional space, efectively reversing the reduction performed by the matrix and the shape
becomes [
Visual instruction tuning [
LLaVA [
To address the caption prediction task, the CS_Morgan team fine-tuned several LMMs that were
pretrained using extensive standard datasets from the field of computer vision. These models were derived
from well-known LLMs commonly utilized in Natural Language Processing (NLP). Ten submissions
were made, and the technical details, methods, and approaches of these submissions are detailed in the
following section. Moreover, the reproducible codes relevant to the following submissions can be found
here [
Before any tasks are performed, the dataset is pre-processed to ensure that it is clean and correctly formatted. Beyond the initial image-text pre-processing described earlier, the training, validation, and testing datasets were structured for generating captions to meet the input requirements of the corresponding vision-language models. Furthermore, the dataset was managed using the Hugging Face Hub. Computational details can be found in Appendix A.
For this submission, the pre-trained LLaVA 1.6 on Mistral 7B weights was loaded using Mistral-7BInstruct-v.0.2 as the base LLM and flash attention was used to optimize attention mechanism computations. To enhance training stability, all float16 instances of the Vision Tower model were replaced with bfloat16. Additionally, prompts were set up by combining images and texts using the "mistral_instruct" conversation mode.
For eficient fine-tuning, LoRA was applied to specific layers, configuring it with a rank r = 16, an alpha (lora_alpha) of 32, and a dropout rate of 0.05. The query, key, and value projection layers in the self-attention mechanisms of the Mistral Decoder Layer, as well as the projection layers in the MLP, were specifically targeted. In the vision model, LoRA was applied to the linear projection layers within the self-attention mechanism (CLIP attention) of the encoder layers in the CLIP encoder. This resulted in 40,108,032 trainable parameters, about 0.527% of the model’s total parameters. The LoRA components included lora_A, lora_B, and lora_dropout representing the low-rank projection to a smaller dimension, projection back to the original dimension, and a parameter to prevent overfitting, respectively.
The training process involved setting up a Data Loader for the dataset, ensuring images and text inputs
were properly loaded. Custom callbacks were defined for printing the best checkpoint and implementing
early stopping. Key training parameters included a learning rate of 1e-4, bfloat16 precision, and the
AdamW [
The second submission was built upon the first one by fine-tuning a larger portion of the model using the same pattern. This included an expanded application of LoRA to improve utilization of the model’s capacity for more accurate and robust predictions. The fine-tuning involved additional layers to enhance the learning and improve visual-textual alignment. Specifically, output projection layers such as o_proj in the Mistral Decoder Layer’s self-attention mechanism and out_proj in the vision model were included to better capture complex relationships within the data, which is essential for tasks like image captioning. Targeting multimodal projector layers (mm_projector.0 and mm_projector.2) enhanced the alignment of visual and textual representations, which is crucial for multimodal tasks. Despite the increased number of trainable parameters (98,467,840 compared to 40,108,032), this expansion represented only a small fraction (1.285%) of the total model parameters, maintaining parameter eficiency while improving learning capabilities. LoRA was configured with a rank r = 32, lora_alpha of 32, and a dropout rate of 0.05. Various layers were targeted in the Mistral Decoder Layers, including query projection (q_proj), key projection (k_proj), value projection (v_proj), and output projection (o_proj) in the self-attention mechanism, as well as gate projection (gate_proj), up projection (up_proj), and down projection (down_proj) in the MLP components. In the CLIP Vision Model, LoRA was applied to the similar projection layers in the attention mechanism and fully connected layers (fc1 and fc2) in the MLP. Additionally, the multimodal projector layers (mm_projector.0 and mm_projector.2) were included to further enhance the model’s capabilities. These modifications were applied to the LLaVA-v.1.6 model and its pre-trained checkpoints on the Mistral-7B.
The training configuration included a learning rate of 1e-5, using the AdamW [
This submission was built on the previous one, maintaining the same general pattern but altering which layers were fine-tuned and the fine-tuning strategy itself. The fine-tuning strategy was multifaceted, employing LoRA to adapt key components, such as attention mechanism projections, MLP components, and multimodal projector layers. Additionally, the language model’s head (lm_head) and the embedding tokens (embed_tokens) were explicitly set as trainable parameters to further enable these parts of the model to learn and adapt to the task. This hybrid approach leveraged the strengths of both LoRA adapters and traditional fine-tuning. Fine-tuning the lm_head allowed the model to better tailor its output generation to specific tasks or datasets, which was particularly important for generating appropriate language or captions from medical images. On the other hand, fine-tuning the embed_tokens layer helped the model learn better representations of input tokens, improving overall performance, especially when the input data distribution difers from the pre-training data.
In this configuration, LoRA was set with a rank r = 32, and the lora_alpha was calculated as 32 × √32 to stabilize training and enhance low-rank adaptation performance. This scaling factor normalized the learning rate for LoRA parameters based on rank, ensuring efective updates without causing gradient explosion or vanishing gradients. A dropout rate of 0.05 was applied to prevent overfitting and maintain generalization ability.
For layers explicitly set as trainable, the lm_head was a linear layer that mapped hidden states to the vocabulary space, generating the final output logits for each token. This layer was crucial for the model’s text generation capability. The embed_tokens layer converted input token indices into dense vectors, providing initial representations of the input tokens essential for the model to process the input text. Both the lm_head and embed_tokens layers had their full weights fine-tuned, in addition to the LoRA adapters.
Overall, this hybrid fine-tuning approach combined LoRA fine-tuning for attention, MLP, and multimodal projection layers with full weight fine-tuning of the lm_head and embed_tokens layers. The total number of trainable parameters was 350,650,368 out of 7,654,729,728 total parameters, making up 4.581% of the parameters.
The training arguments included a learning rate of 1e-5, the AdamW [
For this submission, the pre-trained multimodal language model on checkpoints of Llava v.1.6 Vicuna 7B was loaded, which used the lmsys/vicuna-7b-v1.5 as its base LLM. The model preparation involved configuring LoRA with a rank ( r) of 16, an lora_alpha of 32, and a dropout rate of 0.05. The target modules for LoRA were expanded to include the query (q_proj), key (k_proj), and value (v_proj) projections within the self-attention mechanism of the LLaMA Decoder Layer, as well as the gate (gate_proj), up (up_proj), and down (down_proj) projections in the MLP components of the same layer. Additionally, in the CLIP Vision Model’s CLIP Encoder layers, the key (k_proj), value (v_proj), and query (q_proj) projections, along with the first ( fc1) and second (fc2) fully connected layers of the CLIP MLP, were targeted. Furthermore, the multimodal projector layers (mm_projector.0 and mm_projector.2) were included. This expanded application of LoRA resulted in 34,422,784 trainable parameters out of a total of 7,097,329,664 parameters, constituting approximately 0.485% of the model’s parameters.
The training process involved setting up a Data Loader for the dataset and inspecting batches to ensure correct loading of images and text inputs. Custom callbacks were created for printing the best checkpoint and enabling early stopping. The training used a learning rate of 1e-4, bfloat16 precision, Flash Attention, the AdamW optimizer, batch sizes of 4 per device, and gradient accumulation steps of 8, with evaluation and save steps every 548 steps. The model was trained for a maximum of 10,950 steps (5 epochs), with early stopping configured with a patience of 3 and a threshold of 0.01. The evaluation loss was monitored to select the best model, with lower values being preferable. Early stopping occurred at 4,932 steps, and the best model, saved at 4,384 steps, was loaded at the end. For generating captions during evaluation, parameters included a temperature of 1.0, num_beams set to 1, and a maximum of 512 new tokens.
For this submission, same approach of the third submission was followed. The only diference was the implementation of Vicuna LLM. The total number of trainable parameters is 346,718,208 out of 7,147,481,088 total parameters (4.851% of parameters). The training process employed was similar to the previous submission. The only diference was that the maximum token limit was set to 150 for this submission. The model was trained for a maximum of 10,950 steps (5 epochs), with early stopping configured with a patience of 5 and a threshold of 0.01. Early stopping occurred at 6,576 steps, and the best model, saved at 4,384 steps, was loaded for evaluation.
The LLaVA 1.5 7B shares a similar architecture with that of LLaVA-v.1.6 Vicuna-7B. LLaVA 1.5 checkpoints on 7B parameters were loaded, and the expanded use of LoRA resulted in 84,574,208 trainable parameters out of a total of 7,147,476,992, constituting approximately 1.183% of the model’s parameters. Precision was adjusted from float16 to bfloat16 to enhance computational eficiency, and Flash attention was not enabled in this submission. Instead, LLaMA Scaled Dot-Product Attention (SDPA) was utilized in the 32 layers of the LLaMA Decoder Layer. LoRA was configured with a rank of 32, lora_alpha of 32, and a dropout rate of 0.05. Target modules for LoRA included various projections in LLaMA Decoder Layer, MLP components, and attention mechanisms within CLIP Vision Model. The training process involved creating a Data Loader, defining custom callbacks for early stopping and checkpoint printing, and setting training arguments such as a learning rate of 1e-5, and AdamW optimizer. Training was conducted for a maximum of 8760 steps with early stopping triggered at 4,672 steps, saving the best model. For evaluation, parameters included temperature = 1.0, num_beams = 1, and max_new_tokens = 100.
Moondream2, a small vision language model designed for eficient operation on edge devices, was
evaluated on the ImageCLEF 2024 dataset using pre-trained weights from Huggingface [
LoRA was configured with an alpha ( lora_alpha) of 32, which adjusts the learning rate for lowrank matrices, and a rank (lora_rank) of 64 for the adaptation process. It was applied to specific linear layers in both the vision encoder and the text model. In the vision encoder, LoRA targeted the projection layers (proj), and fully connected layers (fc1 and fc2) within the 27 ViTBlock components. Additionally, LoRA was applied to the fc1 and fc2 layers in the multimodal projection layer, a custom module integrated to adapt the projection layer for the purpose of the caption prediction challenge. In the language model, LoRA targeted the Wqkv, out_proj, fc1, and fc2 layers within the 24 Phi Decoder Layer components. Wqkv in the Phi Decoder Layer represents the combined weights for the self-attention mechanism’s linear projections (query, key, and value). With LoRA applied, the model had 74,422,272 trainable parameters, which was about 3.850% of the total parameters (1,931,904,880).
The training process employed various key parameters and strategies to optimize the model’s performance. The number of image tokens was set to 729, aligning with text tokens. Training spanned 10 epochs over 40,000 steps, using a batch size of 8 and gradient accumulation steps of 4, with evaluation after each epoch. An early stopping mechanism with a patience of 6 epochs and a minimum delta of 0.0001 monitored validation loss to prevent overfitting. Data loading and batching utilized PyTorch’s DataLoader with custom collation for images and text tokens, pre-processed and padded for uniform sequence lengths. Gradient accumulation steps set to 4 simulated a larger batch size for better GPU memory management. The Adam8bit optimizer from the bitsandbytes library, with a dynamic learning rate adjusted via a cosine schedule, was used. Loss computation combined image and text embeddings, processed by the Phi language model. The training loop iterated over epochs and batches, updating parameters post-gradient accumulation and checking validation loss for early stopping. LoRA parameters were optimized with an initial learning rate of 3e-6, scaled by a factor of 4, balancing exploration and convergence. This approach, along with gradient accumulation, enhanced resource use and fine-tuning eficiency.
IDEFICS 9B Instruct [
The lightweight IDEFICS 9B Instruct variant was explored using 4-bit quantization to reduce model size and computational requirements while maintaining performance. BitsandByters (BnB) quantization assigns 4-bit precision to the model using double quantization with the normalized floating-point format (NF4) and bfloat16 precision for computations, crucial for running large language models on smaller devices. For fine-tuning IDEFICS 9B Instruct on the ImageCLEF dataset, the checkpoint HuggingFaceM4/idefics-9b-instruct was specified to load the pre-trained model with 4-bit quantization using the BitsAndBytesConfig class.
LoRA was applied to the query projection (q_proj), key projection (k_proj), and value projection (v_proj) in both the ViT and decoder layers, as well as the perceiver attention and gated cross-attention layers. However, the output projection (o_proj and out_proj) in the decoder, gated cross-attention, and perceiver attention layers did not use LoRA but remained as standard Linear4bit layers. This selective application of LoRA allows for eficient fine-tuning by reducing the number of trainable parameters specifically within the attention mechanisms while leaving other projections, like the output projection layers, unmodified.
Custom callbacks were defined for printing the best checkpoint and early stopping. The training arguments included a learning rate of 1e-4, the AdamW optimizer, batch sizes of 2 per device for training and evaluation, gradient accumulation steps of 8, and evaluation and save steps every 500 steps. The model was trained for a maximum of 8762 steps (2 epochs). Early stopping parameters were set with a patience of 6 and a threshold of 0.001. Evaluation loss was monitored to select the best model, with lower values being better. Early stopping was triggered at 8,000 steps, and the best model, saved at 8,000 steps, was loaded at the end of training.
The Encoder-Decoder model was designed to take an image as input and generate a descriptive caption
as output. In this model, the Encoder was a ViT [
Initially, the pre-trained layers were frozen to focus on training the cross-attention layers. In subsequent epochs, GPT-2 was unfrozen and trained, and in the final few epochs, the ViT was also unfrozen. The Adam optimizer and the One Cycle Learning Rate (OneCycleLR) scheduler are used for optimization. Mixed-precision fp16 training was employed with autocast and GradScaler in PyTorch. The training metrics are cross-entropy loss and perplexity, with both metrics aimed to be minimized. The best model was saved based on validation perplexity and was loaded during caption generation.
The CNN-Transformer fusion model for this submission was built around three core models. First, the
pre-trained ChexNet [
The performance of all the submissions regarding the caption generation task were evaluated using the following metrics.
• BERTScore [
In this year’s evaluation for the ImageCLEF task, BERTScore [
Table 1 shows the results of submissions in terms of the primary metrics of performance. In addition to BERTScore and ROUGE, some other performance metrics were also adopted to assess submission results. These metrics are BLEU-1, BLEURT, METEOR, CIDEr, CLIPScore, RefCLIPScore, ClinicalBLEURT, and MedBERTScore. Table 2 shows the results of the additional performance metrics other than the BERTScore and ROUGE used for the caption prediction task. In both tables, the submissions are listed according to the BERTScore (highest to lowest).
Our results indicate that LMMs, when selectively fine-tuned with fewer parameters, can achieve high performance. Additionally, LMMs obtained through quantization and smaller VLMs can maintain competitive performance in medical image understanding and caption generation. From Tables 1 and 2, it is evident that four diferent submission outperformed the others in terms of the pre-specified performance measurement metrics. Submission 1 using the LLaVA-v1.6-Mistral-7B model with 40.1M ifne-tuned parameters using the LoRA technique achieved the highest scores across several key metrics: BERTScore (0.628059), ROUGE (0.250801), BLEU-1 (0.209298), BLEURT (0.317385), METEOR (0.092682), CIDEr (0.245029), and RefCLIPScore (0.815534). Submission 3, also using the LLaVA-v.1.6-Mistral-7B model with hybrid LoRA fine-tuning approach (350.6M parameters) attained the highest CLIPScore of 0.824171, indicating an improved semantic match between the generated captions and the visual content of the medical images. Submission 10, the CNN-Transformer fusion approach (Pre-trained CheXNet as the encoder and Transformer as the decoder) performs better than other submissions in terms of the ClinicalBEURT score of 0.676905. Finally, the submission 8 which was IDEFICS-9B-Instruct quantized to 4-bit, excelled in capturing relevant biomedical concepts compared to other submissions, achieving the highest MedBERTScore of 0.657460034. Overall, the first submission can be claimed as the top performer because of the highest scores in the primary and secondary metrics. Figure 7 shows the comparison of the submissions in terms of the primary and secondary metrics. The significance of these submissions lies in their demonstration of advanced fine-tuning techniques and model performance optimization in the context of generative models. These findings highlight the evolving landscape of model fine-tuning strategies, advocating for resource-eficient methods that maintain or enhance performance. This is crucial for practical and scalable AI deployments across diverse medical applications. Submission Results for the Caption Prediction Task - Secondary Scores
BLEU-1
In addition to the above-mentioned submissions, Submission 4, utilizing the LLaVA v.1.6 Vicuna 7B with selective fine-tuning using LoRA (34.4M parameters), demonstrated well-balanced performance and closely followed Submission 1. Moreover, submissions 3 and 2, both based on the LLaVA v.1.6 Mistral 7B model but with diferent approaches for optimization, closely followed submission 4 in terms of BERTScore and ROUGE. However, the sixth submission of LLaVA v.1.5 7B, based on another variant of LLaVA, could not outperform the LLaVA v.1.6 variants except for LLaVA v.1.6 Vicuna 7B with hybrid finetuning using LoRA technique (Submission 5). The experiment with the MoonDream2 with 74.4M finetuned parameters in Submission 7 showed competitive performance on the test data relative to the larger models across multiple metrics. Submissions 9 and 10 were based on the pre-trained Transformer-based encoder-decoder models other than the LMMs. VisionGPT2 outperformed the conventional pre-trained CheXNet-Transformer or CNN-Transformer based model in every metric except ClinicalBLEURT. Table 3 shows the generated captions for a test image (ID: ImageCLEFmedical_Caption_2024_test_000016) corresponding to the submissions made for the caption prediction task.
Anteroposterior radiograph of the pelvis showing a sacral fracture (yellow arrows) and a pubic fracture (yellow arrowhead).
X-ray of the pelvis showing bilateral sacroiliitis (yellow arrows) and bilateral pubic symphysis (yellow arrowheads).
X-ray of the pelvis showing a large pelvic mass (arrows).
Plain radiograph of the pelvis showing a large pelvic mass (yellow arrows) with a large right-sided pelvic hematoma.
X-ray of the pelvis showing the presence of a foreign body in the bladder (yellow arrow) and the presence of a foreign body in the rectum.
X-ray of the pelvis showing the fracture of the right pubis.
Anteroposterior radiograph of the pelvis showing a right-sided sacroiliitis.
X-ray of the pelvis showing the fracture of the right ilium (yellow arrows).
CT scan of the chest. The CT scan showed a nodule in the right upper lobe.
Bone defect detected in the axillary region.
This year CS_Morgan team submitted a single submission for the concept detection task. The submission
involved the implementation of ConvMixer [
ConvMixer [
The training process involved developing a ConvMixer model designed for classification or concept detection task with 1,944 unique CUIs. The model was built using TensorFlow and Keras, with key components including an initial rescaling layer, a patch extraction stem, and a series of ConvMixer blocks. The model utilized GELU activations and batch normalization for better performance. The architecture included a global average pooling layer followed by a dense output layer with a sigmoid activation function. Training was conducted over 200 epochs with a batch size of 8, a learning rate of 0.001, and a weight decay of 0.0001. The Adam optimizer was used for training, and the binary cross-entropy loss function was chosen for the multi-label classification task. Performance metrics such as accuracy, precision, recall, and Area Under the Curve (AUC) were tracked during training. However, on the F1-score was reported for the submission. A model checkpoint callback was implemented to save the best model based on validation accuracy. After training, the model was evaluated using the best checkpointed weights.
By implementing this model, the F1-score of 0.107645 was attained on the test data, placing it the ninth position for the concept detection task among the participants. This score indicates that the model’s performance in terms of precision and recall is relatively low, as it represents the harmonic mean of precision and recall, providing a single metric that balances both. The score suggests that the model is struggling to correctly identify and classify the relevant instances among the 1,944 classes, leading to either a high number of false positives, false negatives, or both. This low score reflects room for improvement in the model’s ability to accurately predict the target labels. For a test image (ID: ImageCLEFmedical_Caption_2024_test_000016), the predicted concepts or CUIs based on this ConvMixer model were C0030797, C0000726, and C1306645, whereas the ground truth concepts were C1306645, C0030797, and C0034014 (See Figure 9).
For the Caption Prediction task, submitted models included LLaVA v.1.6 with Mistral 7B and Vicuna 7B checkpoints, as well as the LLaVA v.1.5 7B model. Additionally, a 4-bit quantized instruct variant of the IDEFICS 9B model and MoonDream2, a compact VLM, were explored. Two fine-tuning strategies, selective and hybrid fine-tuning, were utilized. Furthermore, traditional encoder-decoder models like VisionGPT2 and CNN-Transformer architectures were also experimented with. Among these, the top-performing submission was the selective training of the LoRA projectors (40.1M parameters) on the LLaVA 1.6 model with Mistral 7B weights. For the Concept Detection subtask, a single model based on the ConvMixer architecture was submitted, which combines the strengths of CNNs and Transformers.
In future research, the primary aim will be to incorporate Explainable AI and reinforcement learning. Explainable AI will enhance model safety and reliability by identifying potential failures and undesirable actions in LMMs. Reinforcement learning, using context-aware reward modeling, will integrate detailed medical image concepts to improve content understanding and performance in multimodal tasks.
This work was supported by the National Science Foundation (NSF) grant (ID. 2131307) “CISE-MSI: DP: IIS: III: Deep Learning-Based Automated Concept and Caption Generation of Medical Images Towards Developing an Efective Decision Support."
The specifications of the utilised computational resources and environment included two machines. The details are as follows.
• Machine 1 • Machine 2 – Machine Type: a2-highgpu-2g (Accelerator Optimized: 2 NVIDIA Tesla A100 GPUs, 24 vCPUs, 170GB RAM) – GPU: NVIDIA A100-40GB x 2 – Booting Disk: 1000 GB SSD – Data Disk: 1000 GB SSD – Language: Python 3.12.x – Machine Type: n1-highmem-16 (16 vCPUs, 104 GB RAM) – GPU: NVIDIA V100 x 2 – Boot disk: 150 GB SSD – Data disk: 1000 GB SSD – Language: Python 3.12.x – Frameworks: PyTorch 2.x and Tensorflow 2.16.x
[