Metadata-Version: 2.4
Name: sumitENV
Version: 0.1.0
Summary: Run sumit403.p4() in an isolated conda environment without crashing your Jupyter kernel
License: MIT
Keywords: pytorch,jupyter,conda,isolated,environment,sumit403
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.8
Description-Content-Type: text/markdown

# Tenv

Run `sumit403.p4()` in a completely isolated conda environment — your Jupyter kernel **never crashes**.

## The Problem

`sumit403.p4()` injects PyTorch code that conflicts with the standard Anaconda environment, causing the kernel to die every time you run it.

## The Solution

**Tenv** runs `p4()` in a separate subprocess inside a dedicated `p4_env` conda environment. Your main kernel is untouched, figures are captured and displayed inline, and you never need to restart.

## Installation

```bash
pip install Tenv
```

## Usage

### Step 1 — One-time setup

Run this **once** to create the isolated environment with PyTorch + sumit403:

```python
from tenv import setup_env
setup_env()
```

### Step 2 — Run p4 (every time, no crashes)

```python
from tenv import run_p4
run_p4()
```

That's it. Output streams back to your notebook, matplotlib figures display inline, and your kernel stays alive.

## Requirements

- Anaconda or Miniconda (conda must be on PATH)
- Python 3.8+
