| Epoch | \nTraining Loss | \nValidation Loss | \nAccuracy | \nF1 Macro | \n
|---|---|---|---|---|
| 1 | \n0.276373 | \n0.399544 | \n0.798900 | \n0.833400 | \n
| 2 | \n0.254537 | \n0.422038 | \n0.801000 | \n0.835600 | \n
| 3 | \n0.219580 | \n0.442508 | \n0.799600 | \n0.834000 | \n
| 4 | \n0.287847 | \n0.409739 | \n0.798900 | \n0.834600 | \n
"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"Writing model shards: 0%| | 0/1 [00:00","text/html":""},"metadata":{}},{"name":"stdout","text":"\n📊 Final Evaluation:\n eval_loss: 0.4215\n eval_accuracy: 0.8001\n eval_f1_macro: 0.835\n eval_runtime: 54.2312\n eval_samples_per_second: 108.296\n eval_steps_per_second: 1.696\n epoch: 4.0\n\n📋 Classification Report:\n precision recall f1-score support\n\n anxiety 0.94 0.97 0.95 725\n depression 0.70 0.83 0.76 2129\n stress 0.85 0.74 0.79 3019\n\n accuracy 0.80 5873\n macro avg 0.83 0.84 0.83 5873\nweighted avg 0.81 0.80 0.80 5873\n\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Writing model shards: 0%| | 0/1 [00:00 list:\n model.eval()\n out = []\n for text in texts:\n enc = tokenizer(text, return_tensors=\"pt\", truncation=True, max_length=MAX_LEN).to(device)\n with torch.no_grad():\n probs = torch.softmax(model(**enc).logits, dim=-1).squeeze().cpu().numpy()\n out.append({c: round(float(p), 4) for c, p in zip(le.classes_, probs)})\n return out\n\nprint(\"\\nSample prediction:\")\nprint(predict_proba([X_val[0]]))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-04-11T12:58:08.412267Z","iopub.execute_input":"2026-04-11T12:58:08.412999Z","iopub.status.idle":"2026-04-11T14:35:31.904845Z","shell.execute_reply.started":"2026-04-11T12:58:08.412950Z","shell.execute_reply":"2026-04-11T14:35:31.903928Z"}},"outputs":[{"name":"stdout","text":"Device: cuda\n\nClass distribution:\ntarget\nstress 15094\ndepression 10644\nanxiety 3623\nName: count, dtype: int64\nClasses: ['anxiety' 'depression' 'stress']\nTrain: 23488 | Val: 5873\n\nClass weights:\n anxiety: 2.702\n depression: 1.287\n stress: 0.778\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"config.json: 0%| | 0.00/615 [00:00","text/html":"\n
| Epoch | \nTraining Loss | \nValidation Loss | \nAccuracy | \nF1 Macro | \n
|---|---|---|---|---|
| 1 | \n0.437816 | \n0.397521 | \n0.768600 | \n0.807500 | \n
| 2 | \n0.386207 | \n0.406955 | \n0.794000 | \n0.827200 | \n
| 3 | \n0.329776 | \n0.392873 | \n0.793300 | \n0.830900 | \n
| 4 | \n0.270477 | \n0.445115 | \n0.796400 | \n0.830000 | \n
| 5 | \n0.181897 | \n0.511371 | \n0.798400 | \n0.831800 | \n
| 6 | \n0.189214 | \n0.525796 | \n0.797200 | \n0.830900 | \n
"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"Writing model shards: 0%| | 0/1 [00:00","text/html":""},"metadata":{}},{"name":"stdout","text":"\nFinal Evaluation:\n eval_loss: 0.5114\n eval_accuracy: 0.7981\n eval_f1_macro: 0.8316\n eval_runtime: 50.7706\n eval_samples_per_second: 115.677\n eval_steps_per_second: 1.812\n epoch: 6.0\n\nClassification Report:\n precision recall f1-score support\n\n anxiety 0.94 0.96 0.95 725\n depression 0.72 0.77 0.74 2129\n stress 0.82 0.78 0.80 3019\n\n accuracy 0.80 5873\n macro avg 0.83 0.84 0.83 5873\nweighted avg 0.80 0.80 0.80 5873\n\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Writing model shards: 0%| | 0/1 [00:00","application/javascript":"\nvar link = document.createElement('a');\nlink.href = '/kaggle/working/mental_xlmr_final.zip';\nlink.download = 'mental_xlmr_final.zip';\ndocument.body.appendChild(link);\nlink.click();\ndocument.body.removeChild(link);\n"},"metadata":{}}],"execution_count":7}]}